. . .
Vol 34 No 3, July - September 1996 Page 79 PREVIOUS ... CONTENTS ... SEARCH ... NEXT


Regaining Accuracy in a Fluency Oriented Writing Class
by Richard Humphries

The accuracy/fluency dichotomy has provided the EFL/ESL teaching field with a wide arena for both research and disputation. As the trend towards communicative competence continues, with its concomitant concentration on meaning as opposed to form, fluency becomes more of a target for curriculum development and the previous stricter views toward learner errors have been supplanted by methodologies with more relaxed outlooks. Nonetheless, as Savignon (1983) points out, teachers often lack explicit guidelines and have become legitimately concerned with what to correct and when.

With regard to writing skills, Rivers (1968) recognized that sheer accuracy was unreasonable, but countered that without some insistence on accuracy, writing, particularly academic writing, would suffer markedly.

On the other hand, a variety of other researchers, such as Zamel (1987), have stressed the need to emphasize writing in second and foreign language classrooms as a process of discovery, and that an over- emphasis on errors and rhetorical forms can inhibit this process.

For the teacher in an academic EFL/ESL situation, there is something of a dilemma. It is desirable that learners develop effective and creative writing strategies to enable them to use writing as a communicative tool. It is also desirable that they acquire the grammatical and rhetorical forms which they may have to demonstrate competence in, in a future academic or work-related setting. The idea that difficulty with form will correct itself over time is unconvincing to many teachers who do not wish to neglect accuracy completely. Brumfit (1984) suggests a balanced approach, allowing students to produce their own ideas in written form with guidance from the teacher in order to produce a corrected form. He also notes that it is,

"possible to create the conditions for group revision and improvement of written work, so that the accuracy activity is turned into something of a conscious but relatively spontaneous exercise, for talking about accuracy may be fluency talk" (p. 86).

A situation existed with Japanese college-level students in an intensive English writing class that seemed to call for such a balanced approach. Students were making adequate progress at the discourse level, coming to grips with cohesion, organization, and basic rhetorical style, but this was offset by continued and severe problems at the sentence level.

With this in mind, it was decided to implement a class project that would address some of the accuracy concerns but stay within the framework of a curriculum based on the process and communicative approaches to writing. The project took the form of a student-made error correction booklet, similar in some ways to the "Common Mistakes in English" type of publications that are available commercially. The hope was that the learners' sentence-level writing would improve by having them focus on their own errors, with a definite goal in addition to that of revision.

The project consisted of five stages, and was developed over the length of a semester. (Classes met four times a week; lessons were fifty minutes in length.) About 10% of class time was devoted to the project.

Stage 1

Each student was directed to purchase a small notebook, for the purpose of recording his or her own, as well as other classmates' errors during the semester. Examples of commercial grammar-based error correction books were displayed to give them an idea of the direction the project would take. In answer to the question, "Why don't we just buy these?" it was explained that while those books are useful, they often have an artificiality in that they are carefully constructed to provide context and to highlight one error per sentence. Student writing often contains more than one error per sentence. Additionally, most of those books highlight errors of grammar while in student sentences the meaning is often not clear for more contextual reasons (vocabulary choice, words missing, awkward phrasing). The students were told that as this booklet was to be based on their own errors, it was more likely to be indicative of their present and future needs.

Each week the students would record approximately fifteen errors in their notebooks in the following format:

Western clothes is easy to wear, isn't it?

subject-verb agreement/pronoun reference

Western clothes are easy to wear, aren't they?

Most of the errors were drawn from students' weekly journals, with the remainder taken from other class work. Typically, individual students would enter three of their own errors and twelve of the other students' errors in their error notebooks. The teacher at this stage selected the sentences (with variety in mind) and provided the error types as clues to aid in self-correction. Some class time was provided for students to correct the sentences with that information. In areas where a lack of appropriate context inhibited correction, they were encouraged to consult with other students, sometimes searching out the one who wrote the original sentence, in order to have sufficient context. Thus a fluency activity was combined with an accuracy objective. Unfinished correction was assigned as homework.

Stage 2

After four to six weeks, the teacher stopped giving the students information as to error type when the sentences for inclusion were designated. At this point they were able to recognize many of their mistakes for what they were. For those that they weren't able to recognize, other students-and as a last resort the teacher-provided feedback. Eventually this stage could have been extended to the point where students selected their own sentences, but this wasn't done because of time constraints.

Stage 3

After ten weeks the students stopped listing error sentences with corrections in their notebooks. It was now time to produce the class error-correction booklet. All of the students had 150 sentences in their booklets, of which 30 were their own and 120 were drawn from other class members. Taking the class of 16 students as a whole, there were some 600 different sentences containing a variety of error types and combinations. To make the class booklet an effective tool for future writing and to avoid excessive duplication, it was thought that the number of sentences in the booklet should be no more than 100.

A class discussion was held on the significance of errors. In line with research dating back to Burt and Kiparsky's work on global and local errors in the 1970s (cited in Ellis 1990), it was explained that errors that affect the entire meaning, and therefore understanding, of a sentence could be considered more important than others. Generally, the students felt that errors related to verb tense or meaning were more important than those, like articles, which occur frequently but do not often affect meaning.

Groups were formed to discuss the errors in their notebooks for the purpose of assigning each error a ranking based on perceived importance and frequency. The ranking method used was a rank-order scale of 1-5 with 5 being the most frequent/important and 1 being the least important/frequent. A typical notation would be as follows (I = importance; F = frequency):

Recently, I am very busy.

adverbial/verb tense (I-2/F-4)

Recently, I have been very busy.

Stage 4

In groups and individually, students selected the sentences they wanted to include in the final class booklet. These would be the ones they thought to be the most valuable for themselves and for future learners to keep in mind. These sentences were collated and listed. As multiple selection of a particular sentence was often the case, rough averages were taken of the importance and frequency values.

Stage 5

For the sentences selected under the error heading "meaning not clear," contexts were provided by the class in the form of a short explanatory sentence for each . An example of this was as follows:

I am very surprised when I know strange country

verb tense/meaning not clear (I-5/F-1)

Context: The writer had thought the country was better known.

I was very surprised when I heard that country wasn't well known.

The students decided that the final shape of the error notebook would list the errors in descending order of importance. Within each importance level, the sentences would be listed in order of frequency. A short explanatory paragraph about the purpose of the booklet was written by the students and placed at the beginning. The booklet was now finished, and after being typed and photocopied, one was given to each student.

Many might argue that having students choose frequency and importance values for their sentence level errors might produce responses that seem either arbitrary or illogical. Nonetheless, the very act of having students focus on such values with regard to their writing could be seen as a useful end in itself. The students themselves felt that the exercise would benefit the future quality of their written work.

Richard Humphries is currently teaching in the Intensive English Studies program at Kansai Gaidai College in Hirakata City, Japan.



  • Brumfit, C. 1984. Communicative methodology in language teaching: The roles of fluency and accuracy. Cambridge: Cambridge University Press.
  • Ellis, R. I990. Instructed second language acquisition. Oxford: Basil Blackwell Publishers.
  • Rivers, W. M. 1968. Teaching foreign language skills. Chicago: University of Chicago Press.
  • Savignon, S. 1983. Communicative competence: Theory and classroom practice. Reading, Massachusetts: Addison-Wesley.
  • Zamel, V. 1987. Writing: The process of discovering meaning. In Methodology in TESOL. pp. 267-278. Eds. Long, M., and J. Richards. New York: Newbury House.


Excerpts from the Error Correction Book

The information in this class notebook will help you. In this book there are many examples of errors in standard written English. These errors are comparatively frequent ones that Japanese students are apt to make. This book will help your writing. Many students who are learning English are often troubled with grammar. Also, they often write sentences in which the meaning is not clear because the writing method of their mother tongue is different from English. If you can discover your grammatical weak points, you may be able to conquer these frequent errors and master English perfectly.

This notebook was developed and prepared by the 2C2 IES writing class students at Kansai Gaidai, 1994.

Importance Level 5

l. You should study more in left days.

prepositional phrase/post-positional adjective (F4)

You should study more in the days left.

2. My old brother came to my house.

set modifier (F3)

My elder brother came to my house.

3. The mood of the bar was nice with playing jazz music.

word order (F1 )

The mood of the bar was nice with jazz music playing.

4. I am very surprised when I know strange country.

meaning not clear/tense (F1)

Context: The writer thought the country was better known.

I was very surprised when I heard that country wasn't well known.

Importance Level 4

5. So, although we have a little money, we decided to trip to Europe.

meaning not clear (F5)

Context: They didn't have much money for travel.

So, although we have little money, we decided to travel to Europe.

6. It takes about two hours to went to there.

irregular verb preposition (F5)

It took about two hours to go there by car.

7. The camp hold at the youth.

passive/vocabulary (F4)

The camp was held at the youth center.

8. That night was her exciting summer day.

meaning not clear (F4)

Context: She is speaking about one night during summer.

That night was exciting for her.

Importance Level 3

9. It was very hard to study both test.

reword/meaning not clear (F5)

Context: She was going to take two tests.

It was very hard for me to study for both tests.

10. I and she get along with each other

word order (F4)

She and I get along with each other.

11. I have to work to spend a lot of money enough to go to a foreign country.

vocabulary/word order (F4)

l have to work to earn enough money to go to a foreign country.

12. We could see a circus with eating Supain dishes.

spelling/adverbial/article/adjective ( F4 )

We could see the circus while eating Spanish dishes.

Importance Level 2

13. She went to Spain Village her boyfriend.

adjective/preposition (F5 )

She went to Spanish Village with her boyfriend

14. They couldn't enjoy very much

direct object (F5)

They couldn't enjoy themselves very much.

15. My grammar skills should be improved until I take the next one.

vocabulary/adverbial/article (F5)

My grammar skills should be improved by the time I take the next one.

Importance Level 1

16. It's very useful way of learning English.

indefinite article (F5)

It's a very useful way of learning English.

17. I was rather attracted by Audrey Hepburn's beauty and proportion than the movie.

comparison/vocabulary (F5)

I was more attracted by Audrey Hepburn's beauty and figure than by the movie.

Back to Top


Vol 34 No 3, July - September 1996 Page 79 PREVIOUS ... CONTENTS ... SEARCH ... NEXT
. .

On October 1, 1999, the Bureau of Educational and Cultural Affairs will become part of the
U.S. Department of State. Bureau webpages are being updated accordingly. Thank you for your patience.