For the last ten years ELT Journal has been publishing a feature called Key Concepts in ELT. The aim of the feature is to give readers a brief overview of a current topic in ELT and to suggest further reading to follow up some of the ideas presented.

Keith Morrow, current Editor of the journal, revisits one of the published key concepts and suggest some further reading.

In 1993, the communicative approach to language teaching had survived its childhood but was going through the trials and tribulations of adolescence - searching for respectability in the real world. Almost ten years on, has this battle been won? Read on to find out.

A view from 1993
By Tricia Hedge, published in ELT Journal (47/3, 1993)
The term fluency has acquired two rather different meanings in ELT. The first is similar to a typical dictionary entry. For example, 'fluent' is defined by Chambers Concise Dictionary as 'able to speak and write a particular language competently and with ease.' In this meaning it is normally restricted to language production, and in ELT it is normally reserved for speech. It is the ability to link units of speech together with facility and without strain or inappropriate slowness or undue hesitation. Faerch, Haastrup, and Phillipson (1984) include fluency as a component of communicative competence, and define it as 'the speaker's ability to make use of whatever linguistic and pragmatic competence they have.' They distinguish three types of fluency:

• semantic fluency, i.e. linking together propositions and speech acts (also known as coherence);
• lexical-syntactic fluency, i.e. linking together syntactic constituents and words;
• articulatory fluency, i.e. linking together speech segments.

Non-fluency in an English language learner is discernible in frequent pauses, repetitions, and self-corrections, as in this extract from the speech of an elementary learner:

'I enjoy... er... enjoyed... er... making this er... homework... on pronunciation... pronunciation... but... um... you know... I have... there are lots of mistakes so... you see... it helps... it is helping me to... imp (coughs)... make better... my English... '

This learner is beginning to compensate for this non-fluency by using the fillers 'you know', 'you see', in the pauses while he deals with his linguistic uncertainty. It is noticeable, too, that he uses the communication strategy of paraphrase when he fails to produce 'improve' and says 'make better' instead to increase his fluency.

This first meaning of fluency relates to competence in the learner. Course books in the seventies often contained fluency drills aimed at increasing the learner's ability to link syntactic segments with ease. For example, the teacher would set up a chain drill and provide each student with a different prompt in turn which they would have to insert in the correct position syntactically, as in:

Student: I went to the theatre last night. Teacher: (my aunt's house) Student: I went to my aunt's house last night. Teacher: (visited) Student: I visited my aunt's house last night. Teacher: (yesterday) Etc.

More recently teachers have debated whether it is possible to teach gambits, such as fillers, to compensate for fluency.

A second meaning of fluency has developed in relation to the goals of ELT and the nature of classroom activity. Brumfit (1984) argues from a definition of fluency as 'natural language use' and defines the aim of fluency activity in the classroom as to 'develop a pattern of language interaction within the classroom which is as close as possible to that used by competent performers in the mother tongue in normal life.' He lists a set of criteria necessary for achieving fluency activity. These have been simultaneously developed and expanded by a number of other writers and can be summarized as follows:

• the language should be a means to an end, i.e. the focus should be on the meaning and not on the form. Other writers have made similar distinctions e.g. message/medium (Krashen), meaning-focused/form-focused (Ellis).

• the content should be determined by the learner who is speaking or writing.

• there must be a negotiation of meaning between the speakers, i.e. the learners must be involved in interpreting a meaning from what they hear and constructing what to say, not reliant on the teacher or textbook to provide the language.

• the normal processes of listening, reading, speaking, etc. will be in play, e.g. improvising and paraphrasing in speech.

• teacher intervention to correct should be minimal as this distracts from the message.

In Brumfit's view, fluency activities will give students the opportunity to produce and understand items of language which they learn in form-focused work or accuracy work. It is significant that his definition of fluency covers all of the language skills. His suggestions for creating natural language use in the classroom include creative writing, class libraries, and project work.

Much ELT material (e.g. Klippel, 1984) has taken up the concept of fluency activities, and presents tasks which conform to the criteria above. An example would be students reading a handout with five proposals for how to spend a weekend, and agreeing on the weekend they would enjoy spending together. More recently the criteria for task design have been developed in relation to specific skills, e.g. Bygate, 1987, on speaking, and the debate on fluency in the classroom has extended to the roles of teachers and learners and the appropriate management of classroom learning.

Further Reading
Brumfit, C. 1984. Communicative Methodology in Language Teaching: The Roles of Fluency and Accuracy. Cambridge: Cambridge University Press.

Bygate, M. 1987. Speaking. Oxford: Oxford University Press.

Faerch, C., K. Haastrup, and R. Phillipson. 1984. Learner language and language learning. Clevedon: Multilingual Matters 14.

Klippel, F. 1984. Keep Talking: Communicative Fluency Activities for Language Teaching. Cambridge: Cambridge University Press.

A view from 2003
By Keith Morrow
In 1993, the communicative approach to language teaching had survived its childhood but was going through the trials and tribulations of adolescence - searching for respectability in the real world. Almost ten years on, this battle has been more or less won; everybody now claims to be a communicative teacher, even if it still means very different things in practice to different teachers in different classrooms. Perhaps the key feature that unites the different practices in the different contexts is the recognition that a learner who knows the grammar, the vocabulary, and the phonology of a foreign language does not really 'know' the language in any useful sense. Only if s/he is able to make use of this knowledge in order to communicate effectively with other people can the learner truly 'know' the language. Therefore the goal of language teaching in the 21st century is to help learners to acquire this communicative ability - in other words to become 'fluent' in the target language.

One of the signs of the increasing maturity of the communicative approach has been a recent cooling down of arguments about the relative importance of 'fluency' and 'accuracy'. In the aftermath of Brumfit's Communicative Methodology in Language Teaching (1984), some communicative enthusiasts declared that insisting on accuracy was an impediment to learners' development of communicative skills and strategies, and that what mattered was 'getting the message across'; inevitably, this was followed by a backlash in the 90s when grammar was 're-discovered' and re-prioritized. Nowadays mainstream ELT has accepted that both accuracy and fluency are important. Supporters of task-based approaches are careful to identify ways in which specific language features can be drawn out of tasks and focused on in the classroom, while those in favour of grammar-based teaching emphasize the use of authentic material and corpora of 'real' language to connect the language to its communicative purpose.

This is not to suggest that all is quiet on the communicative front. Fluency has now spread outside its earlier exclusive connection with the spoken language, and it is now common to think of the 'fluent' writer or the 'fluent' reader as well as the 'fluent' speaker. Defining what the term means operationally is, however, still a major problem in any of the skill areas, and methodologies for developing fluency are many and varied.

New developments
Let us look round and see what a range of Oxford University Press authors have to say about fluency and how to promote it. Here are six books and one more article from ELT Journal that seem to have interesting things to add to Tricia's Key concept, or that offer practical applications.

1. Success in English Teaching Paul Davies and Eric Pearse (Oxford University Press, 2000)
This new addition to the Oxford Handbooks for Language Teachers series is designed to give relatively inexperienced teachers a practical introduction to language teaching. It provides a rich source of ideas and insights for the practising teacher within a mainstream perspective. In the present context it is interesting to see how the authors recommend different procedures for work focusing on fluency and for that focusing on accuracy.

The main objective of oral accuracy practice is to get learners to begin to produce formally correct versions of new items. In oral fluency practice, the main objective is to get them to use the items in conversations and other communication without hesitation, even if they make mistakes. Their attention should be more on the information they are communicating than on the language. Oral fluency work should continue if possible into free conversations (see Chapter 5), but we will restrict our attention here to fluency practice intended to consolidate the learning of specific language items.

Accuracy practice typically involves using only the new items. Fluency practice, on the other hand, usually combines them with other language. Because of this emphasis on communication and on the more natural use of language, the choice of topics, activities, and types of interaction are even more important in fluency practice than in accuracy practice.

The authors then explore a number of different ways of giving fluency practice including both script-based and unscripted activities. The section concludes with some suggestions for different ways of giving feedback in accuracy-focussed and fluency-focussed work.

Feedback and correction in fluency work
When you decide to intervene to correct errors in fluency practice, the recommended self - peer - teacher correction sequence applies as in accuracy practice. But since fluency is about effective communication without much hesitation, you should not interrupt activities too often. Do so only when many learners are making the same basic error, or when errors interfere with communication. Instead, monitor the activity, note common errors, and deal with them after the activity has finished. One way of doing this is to write sentences with significant errors in them on the board - without saying who made the errors. Then get the learners to identify and correct them.

It is interesting to consider these suggestions in the light of the five characteristics of a fluency activity which Tricia Hedge outlines above, drawing on Brumfit:

• the language should be a means to an end, i.e. the focus should be on the meaning and not on the form. Other writers have made similar distinctions e.g. message/medium (Krashen), meaning-focused/form-focused (Ellis).
• the content should be determined by the learner who is speaking or writing.
• there must be a negotiation of meaning between the speakers, i.e. the learners must be involved in interpreting a meaning from what they hear and constructing what to say, not reliant on the teacher or textbook to provide the language.
• the normal processes of listening, reading, speaking, etc. will be in play, e.g. improvising and paraphrasing in speech.
• teacher intervention to correct should be minimal as this distracts from the message.

The last of these is clearly recognized by these authors. But it is perhaps significant that their starting point is accuracy, and fluency work is seen as a development and follow up, still to be very much controlled and guided by the teacher. This locates their approach very much at the traditional end of the communicative spectrum, though it probably reflects the realities of most language classrooms, and the priorities of most language teachers throughout the world.

2. Individual Freedom in Language Teaching Christopher Brumfit (Oxford University Press, 2001)
As noted in the Key concept, it was Christopher Brumfit's 1984 book which highlighted the distinction between fluency and accuracy in terms of the goals of language teaching. His most recent book, which appears in the Oxford Applied Linguistics series, has an interesting section in which he considers the way in which we, as language teachers or applied linguists, use language to define the categories of our subject (language). He argues that:

'the categories taken for granted at particular times may be the greatest barriers to improvement of practice, for they may act as blinkers, closing relevant options, rather than as telescopes, focusing attention on the most important areas to be concentrated on.' (p.39)

As an example of blockage, he considers the use of the term 'the four skills', and suggests an alternative base for effective language learning. He is also critical of the terms 'presentation stages' and 'practice stages', and of 'controlled', 'guided', and 'free' applied to writing activities. However, he considers the distinction between accuracy ('language work for pedagogic feedback' p.42) and fluency ('language work for an external, real-life purpose' p.42) to be still valid 'because the relative proportions of time and activity spent on each of these allows argument about fundamental views on the nature of language acquisition in direct relation to explicit classroom acts.' (p.42)

'For teachers, whose main applied linguistic problem area is language teaching methodology(as far as page 42)... in direct relation to explicit classroom acts.)'

3. Teaching and Learning in the Language Classroom Tricia Hedge (Oxford University Press, 2000)
Tricia Hedge's new book in the Oxford Handbooks for Language Teachers series is a practical guide informed by, and drawing explicitly on, underlying theories and principles of language learning. On pages 60 - 62 she considers the use of communicative tasks in the classroom and their implications with regard to promoting fluency.

One problem she identifies is the possibility that learners may develop 'undesirable fluency', using convenient but incorrect language to perform tasks. Her main focus is to find ways of integrating activities focusing on accuracy with those focusing on fluency, and she sets out a cycle of preparatory and follow-up activities for tasks to help to achieve this.

She also explores the idea that the balance between accuracy- and fluency-focussed activities might change at different stages in a learner's career, with low level learners doing the greater part of their work emphasising accuracy while more advanced learners concentrate on developing fluency.

The issue of exactly how we might create a link between the two is still the subject of much debate. Clearly, it raises questions about the role of grammar and other elements of the formal language system in the communicative classroom. As suggested earlier, linguistic competence is a fundamental component of communicative language ability and it has perhaps been a misconception among teachers that the communicative approach somehow excuses teachers and learners from a consideration of how to develop high levels of accuracy in the use of grammar, pronunciation, and vocabulary. On the contrary, it is rather a question of how to develop communicative language ability through classroom practice but, at the same time, to ensure an understanding of how language works as a system and to develop an ability to use the system correctly, appropriately, and creatively. (p.61)

4. Writing Christopher Tribble (Oxford University Press, 1996)

5. Grammar Rob Batstone (Oxford University Press, 1994)
These two titles from the Scheme for Teacher Education series are interesting to look at together because although they deal with different content areas, they have similar things to say about the role of fluency in relation to their topics. Although neither author uses the term fluency in their discussion, both identify approaches to their field which draw on the idea of 'process'. Since fluency itself is essentially a process (of making use of the linguistic resources at the language user's disposal), it is interesting to see to see how both authors look at ways of helping this to develop.

For Chris Tribble a process view of writing is contrasted to 'a traditional, text-based approach still in use in many materials' where there is a 'focus on form' (p.37); for Rob Batstone there is a fundamental distinction between teaching grammar as product, and teaching grammar as process, and he devotes a complete section of his book to a demonstration of the latter.

One of his main concerns is to find ways of helping the learner to 'attend simultaneously both to the quality of their language and to the meanings they are expressing'. (p.78) This could be paraphrased as helping the learner to maintain accuracy while developing fluency. He has some interesting suggestions for ways to regulate language use to help learners to focus on different areas, but he emphasises:

We should not have unreasonable expectations of process teaching. Our objective is not to carefully control the learner's accurate production of grammatical forms - this is the domain of product teaching. Rather, it is to develop the skill of exploiting grammar to express meanings as clearly as possible in language use. (p.79)

In other words, to become fluent!

6. Communication and Correction in the Classroom Dermot Murphy, ELT Journal (40/2, 1986, pp 146-151)
One of the major problems for teachers wishing to help their learners to develop fluency is how to handle mistakes their students make. This was raised by Tricia Hedge in the Key concept article and is mentioned in several of the extracts in this feature. Dermot Murphy's article goes back to 1986, and in many ways reflects the concerns of the time to move away from the 'traditional' model of correction where every mistake of accuracy was seized on by the teacher. It is interesting in introducing the idea of 'fluency' mistakes and setting out ways of dealing with these.

Here is the abstract:

In this article I consider the correction of two kinds of error: errors of accuracy and errors of fluency. I describe the way we use feedback to keep talk going in conversation, contrasting this with what often happens in language classrooms. Next I describe the functions of feedback in conversation and exemplify how it is given, concluding with suggestions to develop feedback on fluency in the classroom, so that learners may develop this aspect of language use. The proposals are based on personal experience in the classroom.

7. The Internet Scott Windeatt, David Hardisty, and David Eastment (Oxford University Press, 2000)
This recent addition to the Resource Books for Teachers series contains a range of activities for learners who have access to the Internet, divided into three sections:

1. Core Internet skills
2. Focus on language
3. Focus on skills

Although the label itself is not used by the authors, almost all of these activities are designed to develop fluency as this is defined by the five criteria:

• the content should be determined by the learner who is speaking or writing
• there must be a negotiation of meaning between the speakers, i.e. the learners must be involved in interpreting a meaning from what they hear and constructing what to say, not reliant on the teacher or textbook to provide the language.
• the normal processes of listening, reading, speaking, etc. will be in play, e.g. improvising and paraphrasing in speech.
• teacher intervention to correct should be minimal as this distracts from the message set out in the Key concept. Interestingly, this focus applies to activities in the third section of the book dealing with not only speaking, but also with writing, reading, listening, and integrated skills.