Speaking: The Distance Delta
Speaking: The Distance Delta
Speaking
Summary
In this input we will complete our look at skills with a consideration of speaking. As it is the
last skill to be discussed, you may find what has been said in the previous discussions of skills
applies here as well. In thinking about speaking it will be important for you to incorporate
and apply issues raised, and information given, in the earlier focuses in the course materials
on listening, writing, written discourse and phonology. Ultimately an atomistic approach to
understanding language and skills is unnatural and unrealistic. While we may focus on
separate aspects of language with students, in the end, our underlying aim is to enable
students to be effective communicators or users of English. Current ELT thinking emphasizes
the holistic nature of language and encourages a more holistic understanding and approach
to it. It will be important that you identify and pull together the relevant threads examined
previously in the course of this focus on speaking, in addition to thinking about those issues
particular to speaking.
Objectives
By the end of this input you will have:
Extended the above to a brief consideration of problems students face when speaking
and, later, reasons why fluency activities may not be successful.
Briefly revised ways of providing correction. (See previous input in the course materials
for a thorough discussion of this).
Contents
1. Preliminary Thoughts
2. Generally Speaking
5. Developing Fluency
6. Articulating Aims
7. Error Correction
8. Conclusion
9. Terminology Review
Reading
1. Preliminary Thoughts
Here are some student answers in response to the question, ‘What is most important for
you in your study of English language?’
There can be no doubt that many of our students are motivated by a desire to ‘learn to
speak’ English. They may come to us with a wide discrepancy between their ability to
communicate verbally and their understanding of grammar rules, their ability to write or
read effectively and their ability to use the language to take written tests or to do exercises.
This may be because there are no easily accessible partners to talk with; it may be because
the education system they previously studied in did not promote speaking or because they
studied to pass a written exam; it may also be that they were put off by the ‘academic’
approach to English they have experienced in their previous English studies; you may also
have students who learned more informally from tourists, resident native speakers, etc. and
who have (some) fluency in specific situations. Students are often unclear about what they
really mean by ‘speak’ the language and what speaking the language involves.
Unfortunately, teachers are also often unclear about this, blurring the aims of
communicative practice of particular language items and fluency-focused activities such as
discussion.
2. Generally Speaking
Because the different types of speaking have different characteristics, proficiency in one
event may involve knowledge of specific conventions and/or language different from what is
demanded in another. For example, in an event such as making a speech at a friend's
wedding, speaking will be characterised by planning, scripted anecdotal content, low
reciprocity i.e. little dialogue and set phrases such as ‘I'd like you to raise your glasses…’, ‘I'd
like to make a toast to…’, ‘Could I have your attention please?’, ‘I have been asked to say a
few words about…’. Interestingly, even though the content may be humorous and personal,
the language of the speech itself is often quite formal, and the organisation of the content
prescribed by tradition. In contrast, recounting a personal anecdote to a friend will be
characterised by a lack of planning, higher reciprocity, some set phrases such as ‘It was
[wonderful]!’, ‘We had a [brilliant] time!’, "Have you ever been…?’, ‘You know what I mean?’
and a lack of formality. In other words, speech events belong to different genres, and the
conventions associated with each genre are areas that students and teachers seldom
recognise as crucial aspects of fluency. Please note that conventions are often
generalisations of behaviour, not rigid and inflexible formula.
Look at the different speaking events below. Organise them into groups that seem to share
broadly similar characteristics. Consider, for example issues of formality-informality,
preparation, lack of planning/spontaneity, reciprocity/no reciprocity, and the tenor or
relationship of the interactants (social distance, power, and affective issues).
A news broadcast
Giving a speech
A job interview
A political debate
Making a presentation
There may well be different ways of grouping these speech events, but here is one way of
doing so:
Events: buying theatre tickets, buying fruit and vegetables, meeting someone for the
first time.
In addition to this, there is also particular set phrases associated with these events. These
could include:
Why did you leave your last job? What were your responsibilities?
How do you do? / Nice to meet you. / Sorry, I didn’t catch your name. / How do you
know Sally?
(The events these phrases are used in are clear from the examples)
Many of the events are quite predictably organised. In a speech, there is an introduction and
a conclusion; when meeting someone we often establish certain key personal information
such as name, profession, common interests and connections to other people; when buying
fruit and vegetables we say what and how much of something we want in response to a
question and then conclude by paying and saying thank you. The least predictably organised
events, then, are those which are informal, spontaneous reciprocal and incorporate affective
issues.
It involves the speaker's ability to manipulate language, but not necessarily to produce
only grammatically correct utterances.
It involves interaction with one or more other speakers (we will not be considering any
form of monologue here).
Thus, it necessitates an ability to listen effectively and it is normally reciprocal i.e. the
listener responds in some way to what is said.
It involves the use of gesture, body language and facial expression to clarify and
reinforce the message.
It involves speakers making real-time decisions either about the direction of the
conversation or about dealing with (potential) difficulties in order to ensure the message
intended is communicated.
How information or meaning is conveyed i.e. the structuring of the parts of the event,
the language itself, the phonology
In fact, any one of the above is a potential problem area for students. These general
problems are likely to be articulated as a more specific range of problems students may
have:
Speed of speaking
Grammatical accuracy
Current thinking about learning indicates that students learn what they identify they need.
Thus in a TBL classroom, students are given (or choose) tasks for which there is no pre-
taught language, although there may be phrases presented which will facilitate their
completion of the task. There is sometimes a model task for them to work from and they are
given rehearsal time before their spoken or written presentations. Students' speaking ability
is improved/increased by providing opportunities for them to identify language they need
but don't have and by the teacher challenging them to find other than the safe ways of
expressing themselves wherever possible.
In other words, there has been a shift in ELT from regarding speaking as a means of mainly
or solely grammar and pronunciation practice to recognition of it as the complex medium of
real and meaningful communication.
Before moving on, it is important to briefly consider crucial principles underlying the
Communicative Approach. One of the theories of language underpinning the approach was
formulated by Noam Chomsky. An aspect of his theory of generative grammar was the
notion of ‘competence’, or the theoretical ability that native speakers possess to form
grammatically correct sentences in their language. Dell Hymes (1975) felt that without
including culture and communication in any consideration of competence, the definition
remained abstract and therefore relatively meaningless. He coined the term ‘communicative
competence’ to describe what a native speaker knows about the culture of their speech
community, combined with their (abstract grammatical) knowledge of their first language,
that together allow them to communicate competently. Thus a person's communicative
competence is a combination of theoretical ability to generate grammatically correct
sentences (knowledge) and his/her socio-linguistic competence, i.e. his/her ability to use
language to communicate meaningfully in a culturally appropriate way.
For further reading about the Communicative Approach and the thinking underpinning it,
see Halliday (1970), Brumfit and Johnson (1979), and/or Widdowson (1978).
Traditionally, accuracy is
…judged by the extent to which the learner's output matches some external standard -
traditionally the output of an idealised native speaker [it is judged by] assessing a
learner's command of the linguistic systems, such as grammar, vocabulary and
pronunciation.
To a certain extent our notion of accuracy also includes the factor of appropriacy. We would
probably agree that a student who said ‘I'm desperate for the toilet’ to a member of staff at
an expensive restaurant or who, when talking about today's weather said, ‘It's raining cats
and dogs’ might well be correct but is still wrong or inaccurate in their use of language.
Do you think the terms ‘accuracy’ and ‘fluency’ apply to skills other than speaking? Your
initial response might be negative but it is not unusual for teachers to describe a student as
‘a fluent writer’, that is, one that produces coherent text relatively effortlessly. By the same
token, ‘an accurate listener/writer/reader’ would mean that the student can correctly pick
out specific information from a text and/or that the student does not make many mistakes
in their writing.
Turning now to fluency, here are some statements that provide slightly different
perspectives on the term.
Fluency concerns the learner's capacity to produce language in real time without undue
pausing or hesitation.
Fluency is not so much a matter of the speed of the delivery, but more to do with the
length of the 'run' - the more words you can put together without pausing, the more
fluent you are...the ideal learner is the one who can balance the demands of real-time
processing with the need to be reasonably accurate...They also need to be able to
reorganise (or restructure) what they know in order to make it more complex.
(Thornbury, op cit)
It is our ability to use lexical phrases that helps us speak with fluency. This prefabricated
speech has both the advantages of more efficient retrieval and of permitting speakers
(and hearers) to direct their attention to the larger structure of the discourse, rather than
keeping it narrowly focussed on individual words as they are produced. All this fits very
neatly with the results of computational and language acquisition research.
So as you can see from the quotes above, the current consensus is that simply ensuring
students can manipulate language correctly/accurately is not sufficient in today's classroom;
accurate language use (including certain features of spoken discourse such as cohesion) is an
aspect of fluency, but it does not automatically or necessarily lead to fluency. Fluency is a
more complex issue, defined in different ways by different people. The model of fluency that
seems to be increasingly accepted is that it involves the ability to understand and to
manipulate appropriate language and respond to the message being communicated, in real
time; to follow the speaking and conversational conventions that native speakers of that
language follow; and to use all the above at a level of greater complexity and without undue
hesitation except to think about what to say rather than how to say it. So, is it something we
can ‘teach’, or is it something that just happens as learners make progress? We will return to
this question in a moment.
It is important that the distinctions between the two are clear. Communicative activities
incorporate a range of activities from controlled language practice to freer speaking.
Labelling an activity as communicative indicates only that students communicate with each
other in order to complete the task, or in the process of task completion. An information-gap
activity, for example, may involve students in very controlled language production. While
communicative, it is not fluency-focused.
The term ‘freer activity’ comes to ELT in the context of the Present-Practice-Produce model.
‘Freer speaking’ comes in the production stage, and while it is the stage at which students
are in theory manipulating any language relevant to the task; at which mistakes are
'permitted' to occur; and which is student rather than teacher-centred. It is a stage in a
methodology underpinned by the notion that mastery of language and 'fluency' follow on
from accuracy. In the end then, a freer activity often has an underlying language focus.
And so, we return to the question posed earlier about whether fluency is something we can
teach or does it happen 'naturally' in the course of students studying and learning, without
specific attention paid to developing it?
The answer must be that there are aspects of speaking and conversation that we can
highlight and give students practice of, thereby increasing their awareness and ultimately
their oral fluency. (We shall be looking at activities later). As was stated at the start of the
Spoken Discourse materials, our aim in focusing on speaking is to enable our students to
communicate verbally more effectively and fluently both inside and outside the class; this
means that we need to be clear on what types of speaking our students need/want to be
involved in and specific aspects of speaking and conversation our students find problematic.
Identifying the different genres and the characteristics of those genres enables us to focus
on relevant (sub)skills with students so that students' speaking and conversational abilities
improve.
5. Developing Fluency
We shall now consider ways of developing both fluency in conversation and fluency in
speaking more generally. In the latter case, activities are designed to give students an
opportunity to practise speaking under conditions which are the same as or similar to ‘real
speaking’. Refer to the list above to remind yourself of the characteristics of speaking and so
what these conditions can include. In the first case, activities are aimed at helping students
develop awareness of features and speaking conventions and skills which enable them to
You may consider that all the classroom activities involve language and a degree of fluency
and this, of course, is broadly correct. However, some activities prioritise fluency more than
others. The following is a list of different types of speaking activities that do exactly that.
Roleplay
Problem-solving activities
Story-telling
Ranking activities
Debating a topic
Brainstorming activities
Project work
Simulations
Ordering activities
Instruction-giving activities
Describing
G. Porter Ladousse in Roleplay defines it as an activity in which students assume a role and
are creative and imaginative in playing it. Key aspects of roleplay for Ladousse are: students
create their own reality; they experiment with their knowledge of the real world; they
develop their ability to interact effectively in that world; there are no spectators; and the
environment they are in is safe.
Ladousse makes a distinction between roleplay as an activity for language learning versus an
activity for language practice. Roleplay facilitates learning through doing but an
understanding of what learning entails (often based on a PPP model) must also be
accommodated so that students and teachers feel that learning is taking place.
Roleplay can give students an opportunity to practise or become comfortable with real
world aspects of speaking students such as unpredictability, that is, how to deal with various
unfolding situations, phatic forms of language and other social skills.
It can bring the real world and real experience into the classroom.
It is a flexible activity.
Listening, guiding only when necessary and noting mistakes depending on the class
Organising feedback
Asking questions if there is anything unclear about their role or the activity
Similar to other speaking activities, roleplays can be used for practice of specific target
language. They can be very controlled or freer. They can also be used as an activity to
promote fluency insofar as they help prepare students to deal with the unexpected, to be
aware of the importance of different aspects of conversational register and conventions, and
to make choices in real time. It is crucial that as teachers our aims are clear and that they
underpin our choice of roleplay and its implementation.
Discussion
A successful discussion activity is one which involves all students participating in a
meaningful exchange of ideas, views, etc. It is important that students are well-prepared for
the activity, are motivated by the topic or task and really have something to say about it, and
can learn from the experience. If not, it is hard to describe the discussion or activity as
successful.
There are certain problems that teachers need to anticipate when planning a fluency
speaking focus in their lessons. These could be as follows:
a) Students may not, or may not feel they have anything to say.
h) Students may not see the necessity for speaking fluency practice.
The solutions to these problems are as follows but some of the problems are the similar in
nature and so one solution has been offered for both/all.
a) Choose the topic carefully; ensure the topic is part of an integrated focus, not a one-off
unconnected to other work the students are doing; ensure tasks are clear, focused and
meaningful to the students.
b) As above.
e) How you organise the students i.e. how they are grouped needs careful consideration;
also if students are not used to a fluency focus or opportunities for extended speaking,
implement learner training so that they become increasingly confident.
f) Think about allocating roles in activities; think through who is working with whom; work
out culturally appropriate (and ones you feel comfortable using) strategies for
controlling dominating students e.g. giving quieter students the chance to take on larger
roles or putting stronger students together in one group and quieter ones in a different
group; and encourage a more co-operative approach in a one-to-one tutorial.
h) Ensure aims are clear and clearly communicated to students. You may find that at the
start of a course, you will have to establish a rationale for the balanced approach to
language that you wish to/will take.
There are various other things as well that can make a task easier or harder. These issues
which may be different at different levels and these too can affect the success of an activity:
Too much time pressure can negatively affect a fluency speaking focus. It is important to
allow for not only realistic production time, but also student thinking/planning or
preparation time.
Lower levels cope more easily with more concrete tasks, and higher levels are often
more tolerant of more abstract tasks. For example, asking lower levels to plan a two-day
tour of an interesting city in their country for classmates will be easier than asking them
to agree on the ranking of statements about education.
Similarly, lower levels will speak more confidently and successfully if tasks involve things
they are familiar with and unpredictability is limited, while higher levels can be
motivated with greater unpredictability. The example above illustrates this as well.
Students may have very different opinions and reasons for the order they feel
appropriate in the ranking activity, hence the unpredictability; planning the tour can be
set up to limit the variables and so the unpredictability.
It is important to minimise the linguistic complexity associated with completing the task;
a task dealing with the here and now is linguistically less demanding than one which
involves considering the hypothetical.
The complexity of the task itself can make the task easier or harder. Thus a task involving
groups making decisions about the food to serve at a party they are planning is less
complex and so easier than a roleplay of a negotiation with feed-in cards.
Finally students' familiarity with the topic, regardless of level, will affect the ease and
success with which they complete the task. The more they know about a topic, the more
confidence they will have in doing a task related to it.
As with the problems above, thoughtful and thorough planning, and careful setting up and
student preparation in class for fluency activities can increase the probability of success. In
the end, it is dangerous to assume that fluency activities are easy to do in class; the care that
needs to go into the planning and execution of them should not be underestimated.
Nolasco and Arthur in Conversation define the term as any spoken interaction in which two
or more people, but a limited number, have the right to talk or listen without having to
achieve anything concrete. In other words, everyone can have something to say about
anything. The functions of conversation are primarily:
a) To exchange information
Nolasco and Arthur highlight specific aspects of conversation as potentially problematic for
students in their Introduction (pp7-12), and it is generally agreed that these should be
considered when developing student conversational skills. In the Introduction to
Conversation they identify the following:
The co-operative principle: This incorporates four maxims which contribute to co-
operation in conversation. As a result of these maxims, as native-speakers, we co-
operate in conversation so that we say neither too much nor too little and we try to say
it clearly. Thus, it is imperative that we help students identify how much silence or
thinking time, or how long a turn, is (generally) acceptable.
The creation of meaning: This is how context and the relationship between participants
in a conversation influence what is said. For example, ‘The phone's ringing’ can be
interpreted in different ways. It might be a simple observation as in ‘Oh, listen. That
public phone is ringing. How strange’, a request that someone, probably a friend,
partner, sibling answer it, or an excuse for not doing something else. Students need help
with identifying the speaker's intention beyond the meaning of the words themselves.
Adjacency: This has to do with spoken discourse. It refers to two utterances, one
following the other, where the second is clearly related to the first. Adjacency pairs may
be predictable as in ‘Bye! See you’ ‘Yes, take care’; they may also be less predictable as
in ‘Have you got any money?’ ‘Yes, I've been to the bank’. Many of the more predictable
pairs are incorporated into functional work, but the less predictable ones are rarely
covered.
Turn-taking: This has to do with who speaks, when and for how long. Conversational
participants need to be aware of when someone is finishing their turn and how to join
an on-going conversation or build on what someone else has said, all without undue
hesitation.
Openings and closings: These are the expressions or ways we use to either initiate or
end a conversation, and again they are not necessarily obvious to students.
Topic: There are some topics which are appropriate for conversation and others that are
not; further there are some topics that are more likely to come up depending on the
gender of the participants. An inappropriate topic can result in a dead conversation or
can cause offence. It is important not to neglect raising students' cultural awareness of
what is considered fit for conversation.
Stress and intonation: These two aspects of pronunciation are used to open, maintain
and end conversations; they are also used to signal turn-taking. As such they are
important tools in conversation. Students need help not only in using them
appropriately but also, crucially, in being made aware of them and recognising what
they can be used to communicate.
Gesture and body language: While common to conversation in any language, the
amount of gesture used, which gestures are appropriate, and conventions relating to
body language are culture-specific and misuse of them can cause problems in
communication.
What Nolasco and Arthur have identified are aspects of spoken discourse; they are aspects
of real conversation, however, and a lack of awareness or knowledge of them can cause
problems for students both in terms of understanding what they hear and also in terms of
participating effectively. (Please see the input on Spoken Discourse). There are other aspects
of conversation that we need to sensitise students to as well such as interrupting and
ellipsis. All aspects of managing a conversation are important in enabling students to
become more effective communicators in a conversation.
Nolasco and Arthur identify four different types of activity to do this, to develop students'
conversational ability. They state that:
Although conversational competence can only come from fluency activities or natural
language interaction outside the classroom, there is an argument for the use of
controlled activities which help students develop confidence as well as the ability to
participate in and maintain simple, commonly encountered conversations.
2. It is also important to raise students’ awareness of how native speakers converse. Using
recorded material, students' fluency is developed by focusing their attention on all
aspects of conversation, from significant pronunciation features to the specific language
native speakers employ to initiate, maintain and close the conversation through
observation tasks. Significantly, Nolasco also includes here activities to develop students'
ability to interpret what is happening in a conversation accurately, to develop their
ability to interact effectively and to develop their understanding of culture-specific
features of conversation.
3. Students also need the practice that can be provided in fluency activities. It is crucial
that students get practice in speaking and making decisions in real time, in expressing
their own ideas and opinions, etc., and finally in speaking with a purpose.
4. The final activities are feedback activities. These involve students in analysing their own
and other's success or problems in interaction; they are also to a certain extent,
awareness-raising activities as students may be asked to note for example, strategies
they used to communicate something when they didn't have the specific language
available. As Nolasco says: ‘The objective of feedback is to give students the information
they need to improve on their performance’.
D: Rephrasing
E: Summarising
a) What I mean is
c) So in the end
e) In other words
f) All in all
j) Don't misunderstand me
k) In short
Answers
Clarifying what you mean or correcting yourself: a), d), g), j), l)
Rephrasing: e)
Summarising: f) k)
Can you identify which conversational skills are being highlighted in each of the following
extracts?
Geddes, M. & Sturtridge, G. 1994 Intermediate Conversation MacMillan (Unit 5, p21, Data 1-3)
6. Articulating Aims
As in the case of all other language and skill focuses in this and preceding units, the
importance of clearly, fully, accurately and succinctly articulated aims is vital.
To practise speaking
To provide students with a chance to develop their fluency in a debate about pollution
These, while increasingly detailed, are unfortunately inadequate aims for either a fluency
focus or a focus on interactional skills. Compare them to the following:
To develop student understanding and use of culturally appropriate gesture and body
language used in negotiation.
7. Error Correction
With regard to the development of fluency in speaking, the role of error correction remains
somewhat controversial. What do you think? Should teachers correct student errors in
fluency activities? Which errors? When? How? Why should they correct errors?
a) Correction in a fluency activity should come in a correction slot at the end of the activity.
f) It does not really matter what the student was intending to say if s/he made a mistake in
communicating it.
Here are a few ideas regarding the statements about correction. Generally, it is important
that you identified the extreme position that many of them take, and you have rearticulated
them in a less extreme form.
a) The benefit of doing this is that the teacher does not interrupt the flow of the activity
and keeps the focus between the students. Correction of mistakes can come at the end;
equally, it might come the next day as a warmer, or if the activity has been
recorded/videoed, students could work to identify them/some of them. On the other
hand, the teacher may decide that feedback could more usefully include a focus on
something other than errors i.e. there may be input opportunities as well. Finally the
teacher may have to correct during an activity if communication breaks down.
b) This may well be the students' and teacher's expectation, and so should not be ignored.
Obviously however, peer correction is a valuable tool in the classroom as well.
d) As in (a) above, it is indeed best to try not to interrupt students to correct their mistakes
while they are participating in an activity; the one time it is necessary is if there is a
communication break-down.
e) This raises issues regarding peer correction and how appropriate it is to your students
and the culture in which you are teaching. Generally it is accepted that peer correction is
valuable and also that learners may well need training in identifying errors and providing
peers with feedback on those errors constructively.
f) This is similar to (c). Errors are no longer regarded as a sign of mental laziness or lack of
learning. Instead they are thought to be indicators of learning and so can be positive.
Students should be encouraged to express themselves as correctly as they can;
discounting what a student has said simply because there is an error in the utterance is
inappropriate in today's classroom.
The conclusion that you have probably reached is that we should correct, and that what we
correct has largely to do with our aims; in other words, correction of errors should not be
limited to focusing on a misused tense, a problem with word order or an incorrect
preposition. Correction may focus on any aspect of what the students have said and how
they said it. As with other aspects of ELT and teaching in general, the principles underlying
decisions you make must have been carefully considered.
8. Conclusion
In this section we have looked briefly at speaking and issues connected with it. We have
differentiated between oral (and communicative) practice of target language and
development of students' fluency in speaking. In the discussion, we considered the
characteristics of speaking and different speech events, why speaking is potentially so
difficult for learners and the terms accuracy and fluency. We then looked at activities which
give students an opportunity to practise speaking in real time, purposefully and expressing
their own meaning; we also discussed activities which make students both more aware of
and proficient in conversational skills such as turn-taking, turn-holding, opening and closing
conversation, and creating meaning. It will be important that aims that you write for a
fluency focus are clearly and accurately worded. Finally, we returned briefly to the issue of
error correction and its place in the development of fluency.
9. Terminology Review
Example: The ability to understand and manipulate appropriate language and respond to the
message being communicated in real time.
Answer: Fluency
1. Using structures that are more typical of higher level learners than lower level ones.
2. Two consecutive spoken turns e.g. question and answer, request and acceptance.
For 4 – 7, provide a definition and example for the terms given. There is an example
provided.
Example: Accuracy
4. Transactional language
5. Phatic communcation
6. Back channelling
7. Reformulation
2. Adjacency pair
3. Interactional communication
4. Communication with an aim to doing business or generally getting things done eg ‘How
much do you want for that?’ ‘Can you arrange a meeting for next Tuesday?’
5. Language whose purpose is to smooth the conduct of social relations rather than with a
transactional purpose. This is often formulaic language e.g. small talk; ‘Nice weather
we’re having’.
6. Verbal signals given by the listener to show interest, attention, surprise etc.
7. Students encode the language they know to achieve and outcome and the teacher
reformulates this i.e. makes it more accurate. In other words, the content is dictated by
the student and the form by the teacher.
Reading:
Although not essential to your Module 1 preparation, if you would like to explore this area
further we suggest the following:
Suggested Reading
Bygate, M. 1987 Speaking Oxford University Press
Thornbury, S. July 2000 Accuracy, fluency and complexity English Teaching Professional
Issue 16
Additional Reading
Brown, G. & Yule, G. 1983 Teaching the Spoken Language Cambridge University Press
Carter, R. & McCarthy, M. 1997 Exploring Spoken English Cambridge University Press
Dörnyei, Z. & Thurrell, S. 1992 Conversation and Dialogues in Action Prentice Hall
International
Willis, D. 1996 "Accuracy, fluency and conformity" in Challenge and Change in Language
Teaching (Willis, D. & J. eds) (Heinemann)