Error Correction & SFL

Recently, I’ve been thinking a lot about error correction techniques for speaking classes. One study (Lyster, R. & Ranta, L. 1997, ‘Corrective feedback and learner uptake: Negotiation of form in communicative classrooms’, Studies in Second Language Acquisition 19/1: 37-61) classified correction into six types:

  1. Explicit correction
  2. Recasts
  3. Clarification requests
  4. Metalinguistic feedback
  5. Elicitation
  6. Repetition

As a student of Japanese as a Foreign Language, and living in Japan, I’ve noticed from my own language-learning experience that the first two are by far the most common techniques used and also, as Lyster & Ranta also found, the least effective. The problem is that they focus solely on the form of the utterance. By doing so, I think they relegate, or give the appearance of relegating, the learner’s move into a second-order register. For the learner trying to focus on a primary register of meaning and content this can be frustrating (and possibly slightly demeaning).

Rather than focus on the form of the error then, I’ve recently begun trying to focus on the content as way of developing greater delicacy in the student’s linguistic system. To do this I try to focus on the error as if it were an actual discourse move and from there try to build the range of options available to the student. It is similar to the concept of scaffolding in a Vygoskian sense:

  1. Highlight the error as it relates to meaning;
  2. Negotiate more options of a greater delicacy;
  3. Rephrase

Some examples. One common Ideational problem for Japanese (and many other) EFL learners is not distinguishing between count/mass with mental processes, for example *I like dog. In this case, I ask the student questions like how long they have eaten dog or what it tastes like, focusing on the meaning of the clause, which leads to confusion. Then, I demonstrate the functional difference between dog/dogs and the fact that for count nouns the plural is often the unmarked, general form (chickens are stupid v chicken is delicious) which is the reason for the confusion. Then, we restate the original utterance with greater delicacy, I like dogs.

Another example, this time Interpersonal, is where learners will omit the Finite and Subject in wh-question leading to errors such as what time get up? Here, rather than just providing the correct form, I’ll try to illustrate why it is dis-functional. First, I will get the student to ask the question again and highlight the fact that without the Subject we don’t know who the question is directed at. Then, without the Finite we don’t know the time that the question refers to, today, tomorrow or everyday. In this way, we can build up the elements of the Interpersonal clause.

A final Textual error is a common example of L1 transference from Japanese. When asked about the weekend, some learners will produce *Tomorrow is play tennis. Japanese uses a topic marker, -wa, whereas English uses position. Also the Subject may be omitted if it is understood from context. Thus, the Japanese is Ashita-wa tennis suru (tomorrow-topic tennis will do). First, highlight the error by, perhaps, pretending not to understand. Then, illustrate the difference between I will play tennis tomorrow (Subject as unmarked Theme) and Tomorrow, I will play tennis (Circumstance as marked Theme). Finally, rephrase the original utterence.