A recent discussion about formal versus informal language teaching, on the JALTTALK listserv, led one participant, Charles Adamson, of Miyagi University, to describe the results from his use of Suggestopedia and massive amounts of comprehensible input to help Japanese learners accelerate their English language acquisition process. His action research is from several decades ago, but has major implications for current practice in the context of continued debate over the role of formal instruction in language learning. Professor Adamson expanded his JALTTALK comments in the following article he contributed to the ESL MiniConference newsletter.

In the early 1980s I developed a course based on Suggestopedia and Krashen's ideas as I then understood them. This course offered no grammar instruction whatsoever, but concentrated on making the text (a novel) comprehensible to the students. In keeping with the Suggestopedia model that I learned from Dr. Gabriel Racle, then head of the Canadian Government Suggestopedia project, the students were given a translation to accompany text. Krashen agreed during a discussion that translations can function as comprehensible input and years later re-verified this in an e-mail.

During the elaboration portion of the class, the students were placed in situations where they used the language (oral and written) but received no feedback on their grammar, only on the meanings they expressed. In fact most of the feedback was from other students who could not understand the communication.

At the time the school, Trident School of Languages in Nagoya, Japan, streamed the students, using a modified TOEFL-like test. The students were re-streamed for their second year using the results of the Institutional TOEFL. The students were full time and all received 12 hours a week of English instruction with additional classes in other subjects related to business or liberal arts. The students in the regular program took classes in such things as listening, speaking, reading, writing, grammar, and idioms with a variety of teachers. The students in my course received an integrated four-skills course taught entirely by two teachers.

With the streaming tests serving as pre and post tests, the students in my classes did far better than the ones who received the regular course. At the time Trident was considered to be one of the two or three best schools of its type in Japan, so it was not a question of the quality of the other instruction. By "far better" I mean that the classes were streamed at the beginning of the year and at the end of the year my (and a colleague's) class jumped over the two classes which had been above them. For example, they were class 10 at the beginning of the year but most of them entered class 8 the next year. My special classes were the only ones which consistently had students advance to much higher levels for the second year. One year a student even went from class 10 to class 1, from the lowest to the highest. We also used the method with the highest level students and found similar results. The other teachers, however, complained when we taught the highest students so we began taking the lowest.

Everything about the classes focused on making the messages the students received comprehensible and, to use Krashen's terms, adjusting the filter so that the students wanted to understand. Each week's work was based on a chapter of about 1,500 words. On the first day the students listen to the chapter as the teacher read it twice in the format of a Suggestopedic concert. The first time the students look at both the text and the translation while the teacher reads using special intonations. During the second reading the students just listen. During the remaining classes, the students use the language. At the beginning of the course, this usage consists largely of games, where English is used but the winner is not necessarily the one with the best English. Gradually the activities move to more communicative modes. For example, describing a pictures so that another student can draw it. The conversation activities are structured at first but soon become unstructured and the students spend fairly large amounts of time chatting in English. There are also role-plays based on scenes from the novel.

Many people argue that formal instruction is necessary, but in this course none is given and the results are superior to the courses where it is used. Many people who insist on formal instruction point to the situation in Canada to prove their point. However, the teachers in the Canadian program apparently made no attempt to insure that the students understood the input they were receiving. My own work has made it abundantly clear that it is crucial that the students understand the input. They must use it to get reactions from others and they must modify it to express other ideas. The input must be massive and both comprehensible and comprehended. The Canadian situation is typical of much of the work in our field - the researchers accept or reject a method on the basis of a study that does not actually include the key points of the method.

Just to be absolutely clear, I am not so much supporting Krashen's various ideas as supporting the one central idea that understanding the linguistic input is sufficient and no grammatical instruction is necessary. However, this only applies if the students received a sufficient amount of input. The typical commercial textbook does not contain a sufficient amount of text for the students to acquire the language. In my classes mentioned above, over the year the students had about 45,000 words of comprehensible input, plus whatever they received from the teacher and the other students. I know that the text was comprehensible because the students got a translation. Also the teacher concentrated on making classroom talk comprehensible and observing student reactions showing comprehension.

Article by Charles Adamson
adamson@myu.ac.jp
Miyagi University

2004 ESL MiniConference Online