Вы находитесь на странице: 1из 152

Masaryk University Faculty of Arts

Department of English and American Studies

English Language and Literature and Teaching English Language and Literature for Secondary Schools

Bc. Petra Šolcová

Teaching Speaking Skills

Masters Diploma Thesis

Supervisor: James Edward Thomas, M. A .


I declare that I have worked on this thesis independently, using only the primary and secondary sources listed in the bibliography.



Author‟s signature

Acknowledgement I would like to thank my supervisor James Edward Thomas for his guidance, insightful comments and suggestions.

This project was funded by the Action Austria/Czech Republic grant.


Table of Contents

1 Introduction




Thesis structure



Underlying motivation and research questions



Target age groups of learners



What is „teaching speaking skills‟?


2 Theoretical framework


2.1 Basic features of speaking




On the seemingly disorganised nature of speaking


2.1.2 Classroom implications: teachers‟ expectations


2.1.3 The role of context


2.1.4 Classroom implications: teaching materials


2.1.5 The types of communicative exchanges


2.1.6 Classroom implications: communicative tasks use


2.1.7 Characteristics of conversation


2.1.8 Classroom implications: challenges of conversational classes


2.2 Communicative competence



2.2.1 Historical background


2.2.2 Individual components of communicative competence


3 Practical part



3.1 Individual aspects of teaching speaking skills



3.1.1 Fluency and accuracy


3.1.2 Corrective feedback and evaluation


3.1.3 The importance of pair work in LT


3.2 Practical techniques




Information gap activities



Cooperative activities






Storytelling and retelling



Role plays and simulations



Discussions and debates




Creative tasks





4 Research


4.1 Introduction



4.2 Data analysis



Teaching approaches



Communication and the success of teaching speaking skills



Teaching materials and using transcripts of recordings



Accuracy versus fluency and corrective feedback



Assumptions about teaching speaking



Practical techniques for teaching speaking


4.3 Conclusion







Appendix A List of abbreviations








Záložka není definována.

Appendix B Glossary of terms

Záložka není definována.

Appendix C Activities

Záložka není definována.

Appendix D Questionnaire

Záložka není definována.

Appendix E Charts and tables

Záložka není definována.

Czech resume

Záložka není definována.

English resume

Záložka není definována.


1 Introduction

This thesis sets out to show the full scope of teaching speaking skills, which in

its entirety means more than just teaching „speaking‟. The concept of teaching

speaking skills is presented with respect to the notion of communicative

competence. The main line of argument is consistent with the view that in

order to become competent speakers, learners must acquire a good command

of all fundamental areas of communicative competence: linguistic,

sociolinguistic, discourse and strategic.

In addition, this thesis considers the differences between speaking

and writing and advocates a teaching approach which takes these differences

into account. It also attempts to link relevant theoretical concepts and thoughts

to practice and considers their implications for language teaching.

Furthermore, it aims to present some of the ways of practising

speaking skills in the classroom focusing on fluency and offers a variety of

practical techniques which can be used in the classroom. Finally, it seeks to

answer my initial research questions, which are presented in one of the

following sections of this chapter.

The research part of this thesis consists of a comparative study that

investigates how teaching speaking skills is realised in practice by teachers in

the Czech Republic and Austria. Moreover, primary focus is placed on the

teachers‟ assumptions and possible misconceptions about speaking as well as

the classroom techniques they use.


1.1 Thesis structure

This thesis is divided into four main parts. In the introductory part the

underlying motivation behind the choice of the particular topic is discussed and

my research questions and hypotheses are presented. There is also a

subsection providing an explanation why the focus of the thesis is not restricted

to any particular age group of learners and a section attempting to define the

concept of „teaching speaking skills‟.

The second part explores the subject matter from a theoretical

perspective linking the theoretical aspects to practical implications for classroom

teaching. First of all, it deals with the basic features of speaking, such as its

different organization in comparison with writing and the role of context. It also

makes a distinction between two types of communicative exchanges and

discusses the basic characteristics of conversation. It further presents the

concept of communicative competence, describes its components and discusses

the individual aspects which they entail.

The third part is practical and aims to present a range of

communicative activities that can be used in speaking skills lessons. Apart from

this, it discusses several individual aspects of speaking skills teaching, in

particular the fluency/accuracy dichotomy, corrective feedback and the

importance of pair work in language teaching.

The final part presents my own research in the area and suggests

further directions for investigation. The appendices include a list of

abbreviations, a glossary of terms, samples of activities to practise speaking


skills, a questionnaire used in the quantitative research and charts and tables

accompanying the research part of this thesis.

1.2 Underlying motivation and research questions

In my profession as a teacher and as the head of a private language school, I

meet hundreds of students seeking language training each year, mainly adult

learners. In the introductory conversations we have with newcomers the

majority of them stress that they especially need and want to improve their

spoken language production. Moreover, a substantial proportion of people

consider themselves to be very poor speakers.

Interestingly, although they may be coming from different

backgrounds and their motivations for language study may differ be it school,

work or personal interests almost all of the potential learners agree that what

they primarily want to practise is speaking. They also wish to gain practical

skills which they could utilise in their lives.

Upon hearing the often repeated requests for practical speaking skills

training, one naturally starts to ask oneself why it is, that the need for learners

to practise speaking is so great.

It is undisputable that face-to-face communication is the most

fundamental mode of human language” (Givón 1997: 92) and its training is

naturally sought out by learners for a number of obvious reasons. On the other

hand, this does not explain why so many learners frequently feel there is a lack

of speaking skills practice in schools and courses.


Even though learners are to a certain extent responsible for their own

learning success, or the lack thereof, teachers can greatly influence their

learning experience and language acquisition (LA). After all, teachers are the

key players in the way lessons are organised and what skills are taught, down

to the individual tasks that students deal with. According to Tomlinson,

researchers stress that it is the teacher who determines what actually happens

in the classroom. Referring to Williams and Burden (1997) he further adds that

“the teachers‟ own beliefs can affect classroom action more than a particular

methodology or coursebook(Tomlinson 2008).

Although some teachers might argue at this point that they need to

follow certain guidelines or that they do not have much of a say when it comes

to decisions about coursebooks it cannot be denied that they themselves opt for

inclusion or exclusion of certain activities and extra materials in their lessons.

Furthermore, they also decide on the way these resources are used and how

each individual task is put into practice. Even under very restricted conditions,

this gives them some leeway to shape the nature of their lessons on a daily

basis and to place emphasis on the subject matter they consider important.

My hypothesis is that although many teachers may think that

teaching speaking skills is an important part of a curriculum, their assumptions

about speaking may prevent them from being successful in teaching speaking. I

suspect that many teachers fail to recognise important differences between the

nature of spoken and written texts in the first place. As a result, their speaking

skills training fails to meet their learners‟ needs because it does not reflect real-


life conditions of speech production and the scope of the skills that are

necessary for successful communication.

As early as in 1979, Allwright reports that “language teaching,

globally, has not led to a satisfactory level of communicative skill in the vast

majority of cases” and in making a reference to the content of textbooks and

national syllabuses he says that there has been an “apparent failure to ensure

that communicative skill is adequately represented in language courses”

(Allwright 1979: 167). Twenty-six years later, Thornbury still suggests that one

of the reasons why students complain about inefficiency of language teaching

(LT) when it comes to speaking is lack of genuine speaking activities

(Thornbury 2005: 28).

In my research I would like to put this hypothesis to the test. My

research supports Thornbury‟s view that students need an abundance of

practical skills training opportunities and interactive speaking itself.

Furthermore, I assume that in general, teachers do not succeed in providing

these in their lessons on a broader scale. If this hypothesis proves true, one of

my objectives is to examine the reasons why this might be happening.

There are several different scenarios that might emerge from my

research. I hope to find some indication which of the following hypotheses is

the most probable to apply in reality.

My initial research questions are:

a) Do teachers deliberately omit speaking skills practice from their

tuition and if so what is their justification?


b) Do teachers think they teach speaking skills but due to various

reasons teach something else instead or fail to interpret their training

as speaking skills training to their students? Here I would also like to

look at teachers‟ possible misconceptions about speaking.

c) Do teachers think they include enough speaking skills practice but

this is not sufficient for learners or the teachers‟ choice of activities

does not meet their learnersneeds? Here I would also like to find out

what the teachers‟ self-image is in terms of their success in mediating

speaking skills.

d) Do teachers devote sufficient time to teaching speaking skills in a

meaningful way and exploit variety of activities but it is the students‟

perceptions and their beliefs about the ways languages should be

taught that are problematic? Here I plan to investigate the teachers‟

preferences for activities they use to teach speaking.

In order to find answers to some of the above-mentioned questions I

plan to carry out a survey in the form of online questionnaires that will be

distributed to Czech and Austrian teachers of English. My choice to conduct a

short comparative study of two countries can be partly explained by the belief

that I hope to be able to see if there are any differences between the two

countries. The potential similarities and differences would be interesting firstly

because of the historical link that has existed between the two countries for a

long time and secondly because the fact of having data from two countries at

my disposal would allow me to choose a contrastive approach.


Contrasting the two countries might help me to interpret the results

more precisely. For example, if the research showed that the teachers in one of

the countries considered themselves more successful in teaching speaking

skills, it would be interesting to examine if this belief originates in the ways they

teach. My next step therefore would be to compare the activities and strategies

these teachers use with those used in the other country. My objective would be

to find out if any distinctions in terms of different teaching techniques and

assumptions about LT can be identified in the first place and if they can, to

what extent they might be linked to the potential success or failure of learners

in acquiring speaking skills.

Another point I would like to make here is that the importance of my

research can be substantiated by two major facts. Firstly, it is the recent rise in

interest in the area of speaking skills teaching. Secondly, it is the high number

of language educators dealing with teaching speaking skills in their language

tuition on a daily basis. Both these facts make speaking skills research highly

significant. In addition, as far as the former is concerned, even though the

interest in this area can be observed over the last thirty years or so, the

intensity of this interest has recently increased greatly. Here I am especially

referring to the last few years when speaking and its teaching finally started to

get the scope of attention they truly deserve with books by authors such as

Thornbury (2005), Nation and Newton (2009), Hughes (2011), etc.

Furthermore, I hope that the questions under investigation could

throw some more light on teachers‟ current attitudes towards teaching speaking


skills and their ways of dealing with them in the classroom environment seen

from a Central-European perspective.

This particular perspective is essential. EFL contexts are generally

under-researched when compared to ESL and there is a growing demand for

research data from countries which have traditionally been disregarded in

literature on methodology and LA (Tomlinson 2008). Apart from filling the

present gap, this research might also open new interesting questions and

indicate directions for further investigation.

Finally and very importantly, the overall contribution of this thesis

might also lie in its attempt to link theoretical concepts to their practical

implications for LT because as Cook (2009) points out: “despite the vast strides

in SLA research, few people have been interested in maintaining the bridge to

language teaching and only a fraction of the research has been applied to

classroom teaching”.

Consistent with this view is my last point that if the primary goal of

theoretical research did not consist in its further application in practice, the

research itself would become meaningless. This thesis therefore seeks to

consider practical implications of first language acquisition (FLA) and second

language acquisition (SLA) research for LT, linking theory to practice.

1.3 Target age groups of learners

This thesis does not restrict its focus to any particular age group for several



Firstly, many aspects of both foreign and second language acquisition

still remain unclear and need to be explored further including the question of

language acquisition (LA) at certain stages of life.

Unfortunately, there has been no systematic treatment of language

acquisition within the age group of adult learners in an EFL context in modern

books dealing with language teaching methodology or LA, and where the

question of age in LA is frequently discussed (e.g. Lightbown and Spada 2006,

Brown 2008, Hoff 2009, Hughes 2011, etc.). It is precisely this target group I

would be most interested to investigate. The fact that this field remains still

under-researched (Bygate 2009) would push this thesis beyond its scope, if the

target group was specified.

Secondly, there is no unified opinion on some issues of L2 teaching,

namely its relation to FLA. In general, it is now understood that certain

processes of FLA and SLA cannot be the same, 1 partly because learners already

have their knowledge of L1 which they build on. Correspondingly, language

transfer is one of the processes that distinguish FLA from SLA and influence it

both positively and negatively.

On the other hand, it is also true that scholars believe that many

aspects 2 which play an important role in FLA are likely to play a part in SLA, too.

Nevertheless, it remains unclear to what extent FLA processes can be

successfully transferred into L2 teaching and learning. Additionally, which

1 e.g. cognitive development (see Lightbown and Spada 2006)

2 e.g. input and output or exposure to modified input (Lightbown and Spada 2006)


factors might facilitate or impede the transfer is also yet to be ascertained.

(based on Thornbury 2007: Chapter 7; Clark 2000)

As can be seen, with so many areas yet to be explored and questions

still to be answered it would be difficult to draw distinctions between different

age groups. Restricting my references to one specific target group might place

this thesis on fairly unstable ground since there is a considerable amount of

research yet to be done especially as far as adult learners are concerned.

To deal with these gaps in the present understanding of FLA and

SLA, this paper makes references to the findings of both FLA and SLA research,

which can be considered as relevant for the purposes of this paper on the

following grounds.

First of all, the extent to which FLA and SLA differ is still a subject of

investigation, as suggested above. Secondly, it is generally believed that there

is some correlation between child LA and SLA or foreign LA respectively, even

though there is no consensus in terms of the applicability of L1 principles on

EFL teaching. Thirdly, L2 research frequently arises from L1 research findings,

so FLA research will always offer new perspectives to SLA and foreign language

teaching, opening new questions to be answered. In fact, this makes all FLA

research relevant for those dealing with SLA, as the former is highly suggestive.

Lastly, not only does FLA research raise important questions for further

investigation but, it may also have important implications for teaching from the

perspective of a teacher which are to be considered.

In conclusion, coming back to my choice not to limit this thesis in

terms of a specific target group of learners, I would also like to hope that the


thesis might be of some interest to anyone involved in teaching English who

happens to read this text, regardless of age groups they teach 3 .

1.4 What is teaching speaking skills’?

Rebecca Hughes makes an interesting methodological point that as far as

teaching speaking skills is concerned one needs to distinguish between

„teaching the spoken form of a language‟ and „teaching a language through

speaking‟. She also stresses the fact that unfortunately, when compared to

writing, the spoken form is under-researched and that this may be one of the

reasons why teachers may feel more confident when using „stable written forms

and genresin their lessons (Hughes 2011).

At this point, I would like to argue that both of the above-mentioned

concepts are interconnected. This may also be one of the reasons why they are

rarely distinguished from each other when educators speak of teaching

speaking. To explain, I have observed that teaching the spoken form of a

language is not very useful if it is not practised through speaking. By analogy, it

can be argued that teaching speaking if the data used comes from written

genres cannot bring much of a result in terms of progress in spoken fluency

either. Therefore teaching the spoken form of a language using samples of

spoken texts should be part of teaching speaking.

On the same subject, I believe that Rebecca Hughes‟s statement that

teachers may incline toward the written form of a language when teaching

3 Having said that, it needs to be taken into account that my perspective of reasoning, choice of resources, activities, etc. are bound to be influenced by my ten years experience teaching adult learners.


speaking is consistent with my main line of argument. That is, that one of the

problems with teaching speaking skills in the traditional classroom setting is

that it is not the spoken but the written form of language and its characteristics

that are taught. As a result, teachers do not meet their students‟ needs when it

comes to speaking skills training because in the end it is not speaking skills that

are taught.

What exactly is it then that the terms speaking skills and speaking

skills teaching refer to? The definitions of both these terms are closely knit

together with the definition of speaking.

Speaking has often been narrowly defined. When speaking skills are

discussed, this often happens in a context of public speaking. Speaking,

however, is much more than that. Broader views focus either on communication

realised to achieve specific purposes, e.g. to inform, to ask for explanations,

etc., or they describe speaking in terms of its basic competences used in daily

communication such as booking a room, giving directions, etc.

What these approaches have in common is that they view

communication and speaking as an interactive process in which individuals

alternate in their roles as speakers and listeners and employ both verbal and

non-verbal means to reach their communicative goals. Chaney‟s definition

describes speaking in a similar way saying that speaking is the process of

building and sharing meaning through the use of verbal and non-verbal

symbols, in a variety of contexts” (Chaney cited in Kayi 2006).


Consistent with this view, is Nunan‟s description of what teaching

speaking involves. According to him, to teach speaking means to teach

language learners to:

Produce the English speech sounds and sound patternshim, to teach speaking means to teach language learners to: Use word and sentence stress, intonation

Use word and sentence stress, intonation patterns and the rhythm of the second language.to: Produce the English speech sounds and sound patterns Select appropriate words and sentences according to

Select appropriate words and sentences according to the proper social setting, audience, situation and subject matter.intonation patterns and the rhythm of the second language. Organize their thoughts in a meaningful and

Organize their thoughts in a meaningful and logical sequence.social setting, audience, situation and subject matter. Use language as a means of expressing values and

Use language as a means of expressing values and judgments.their thoughts in a meaningful and logical sequence. Use the language quickly and confidently with few

Use the language quickly and confidently with few unnatural pauses, which is called as fluency. (Nunan 2003)Use language as a means of expressing values and judgments. Therefore, whenever the terms speaking skills

Therefore, whenever the terms speaking skills and teaching speaking

skills are mentioned in this thesis, they refer to all the above-listed aspects. It

needs to be pointed out, however, that the scope of this paper does not allow a

special focus on teaching phonetics and phonology. Since this topic can easily

be singled out and treated in a separate paper, the first two above-mentioned

points will be excluded from further discussion. Nevertheless, it is clear that

teaching speaking skills involves teaching these features as well.

Before closing this chapter, I would like to make a distinction

between another two terms that are frequently mentioned throughout this

thesis: language acquisition and language learning, a distinction that has been

made since Krashen and Terrell (1983) and their acquisition-learning

hypothesis. They say that “language acquisition is the „natural‟ way to develop


linguistic ability, and is a subconscious process”, which is realised through

communication. This means that people are “not necessarily aware that they

are acquiring language(1983: 26).

By contrast, language learning is „knowing about‟ language or „formal

knowledge‟ of a language. Learning is therefore conscious and refers to “explicit

knowledge of rules, being aware of them and being able to talk about them”.

(p. 26).

Similarly to Krashen and Terrell, this thesis builds on the view that

second-language learners, regardless of their age, can acquire languages

through communication and meaningful activities. However, this view is

somewhat reserved with respect to the evidence that shows that certain

features of language “seem to respond only, or better, to instruction”

(Thornbury and Slade 2007: 231). Thus effective LT is such where teachers are

able to a) create the right conditions for LA through giving learners

opportunities to experience natural language use and b) provide formal

instruction with features that are better or only learned through instruction.


2 Theoretical framework

2.1 Basic features of speaking

Before some ways of teaching speaking can be discussed and compared to

practices of teachers in the Czech Republic and Austria, it is important to

understand what the main features of speaking are, as opposed to writing for

example, and what skills are involved in the ability to speak a second language.

These are the main points presented in this chapter.

2.1.1 On the seemingly disorganised nature of speaking

One of the basic features of speaking is that it takes place in real time. Due to

the time constraints that allow speakers only limited planning time, speech

production requires real-time processing(Thornbury 2005: 2). This is one of

the main reasons why language learners, and this is not only restricted to true

beginners, tend to find speaking difficult. Strategies used to „buy planning-time‟

(Thornbury and Slade 2007) significantly shape the nature of speaking and

distinguish it from writing. Crystal and Davy also mention time as the main

factor which distinguishes written from spoken language(1979: 87).

Consequently, „instances of disfluency‟ (Thornbury and Slade 2007)

like hesitations, word repetitions, false starts, unfinished utterances and repairs

make speaking look less neat and tidy than writing when transcribed. Therefore

one might assume that speaking is disorganised or even inferior to writing. But

this is not true in reality. As Halliday explains “the formlessness of speech is an

artefact of the transcription”. If judged from the perspective of written texts,


spoken language will always look chaotic on paper because first and foremost,

“it wasn‟t meant to be written down. (Halliday 1989: 77)

Burns and Joyce also state that [


speech, far from being

disorganised, has its own systematic patterns and structures they are just

somewhat different from those in written language” (Burns and Joyce 1997: 7).

For this reason, judging speech through the measures of writing means to deny

its basic characteristics and the purpose for which it is used. Likewise if written

texts like contracts, articles and reviews are rendered in spoken form, they also

sound unnatural. Moreover, they are difficult for listeners to follow because

they were originally created to be used for a different purpose and through a

different channel (Crystal and Davy 1979). An analogy, albeit inverse, is also

true for spoken texts.

As early as in 1989, Halliday argues that “the spoken language is

every bit as highly organised as the written, and is capable of just as great a

degree of complexity. Only it is complex in a different way.(Halliday, 1989:

87) In the book Spoken and Written Language, he also suggests that unlike

written texts, spoken texts are dynamic and tend to have a lower degree of

lexical density. By dynamics he means the tendency of spoken language to

represent experience as processes. According to his words, written language

describes the world in terms of its products and makes a greater use of

nominalization, e.g. improvement instead of improve. By contrast, when one

talks, one says that “something happened or something was done” (1989: 81)

and one therefore tends to use more verbs.


In reference to lexico-grammatical structures, Halliday states that

written texts are lexically dense and their sentence grammar is simple, whereas

spoken texts having a lower level of lexical density have a greater degree of

grammatical intricacy. In conversation, however, this intricacy may be realised

across turns.

Illustrating his view on a number of examples, he further explains

that by grammatical intricacy he means the tendency to use a broader variety

of tenses and aspects in speaking. According to him, spoken discourse makes a

frequent use of clauses and employs a greater variety of both syntactic and

semantic relationships. He says that “it is often thought that sequences of

conversational discourse [] are simply strings of „ands‟. [] Rather, they are

intricate constructions of clauses, varying not only in the kind of

interdependency (parataxis or hypotaxis) but also in the logical semantic

relationships involved.” (1989: 86)

On the same subject, in particular syntactic relationships in spoken

texts, Crystal and Davy state that “the most obvious continuity feature is simple

addition of another structure, itself grammatically independent, using a

conjunction” (Crystal and Davy 1979: 88), a view that Halliday seems to find

too simplistic as suggested above.

More than two decades later and building on the present research

evidence, Thornbury reports that the „grammaring‟ of spoken texts is

“constrained by how much information can be held in working memory at any

one time” (Thornbury 2005: 4). As a result and as a way to compensate for

limited planning time, speakers use the so-called „add-on strategy‟ in places


where written texts might use embedding or subordination. This strategy

means that utterances, phrases or clauses, are added one after another and

glued together by the insertion of the appropriate grammatical markerslike

articles, auxiliary verbs and word endings (2005: 4). Thornbury therefore

describes speech as not only spontaneous but also as essentially linear‟. By

linearity he means the fact that speech is produced utterance-by-utterance.

Thornbury and Slade also talk about “layering of phrase on phrase rather than

forming sentence by sentence as in written texts(2007: 13).

This linear aspect of spoken texts is also dealt with in detail in A

Grammar of Speech by David Brazil (1995). In this highly interesting and

innovative book, the author analyses grammatical, syntactical, discoursal and

intonation patterns that arise from the real-time processing demands and the

interactional character of spoken language. Expressing his dissatisfaction with

the poor representation of speech in conventional grammars, he suggests a

new way of looking at the structure of spoken grammar. Rather than thinking

of words and other entities as occurring at places in a hierarchically arranged

structure”, he analyses spoken language in terms of “chains of elements

occurring in time” (1995: 47). On a systematic basis, he thus manages to

demonstrate a clear structure of a spoken grammar dealing with its individual

elements. Since this is rather a unique work, it would be desirable if more

studies into the nature of spoken language emerged and also if the link

between spoken data and speaking skills teaching was further explored.

Going back to the subject of lexical density mentioned earlier in this

section, advocating Halliday‟s notion, Thornbury and Slade say that “another


characteristic of spoken language which is attributable to its spontaneity is the

fact that information is relatively loosely packed.” (Thornbury and Slade 2007:

13) Therefore both informational (as in Thornbury and Slade 2007) and lexical

sparsity (as in Halliday 1989) of spoken texts can be explained by the real-time

processing and the time constraints speakers need to deal with.

In addition, it can also be observed that both these aspects of spoken

language are inevitable from the listeners‟ perspective as well because while

written texts allow readers to read a text as many times as necessary, spoken

texts do not (Burns and Joyce 1997). Consequently, listeners need enough time

to process the content of the utterances. Should it be too packed with

information or lexis, listeners would find it hard to absorb all that is being said.

Moreover, should the situation also require interaction from their part, it would

be quite likely for such a conversational exchange to shatter.

In summary, the aim of this section was to show that spoken

language “is structurally patterned, and displays an orderliness that is neither

chaotic nor random but, rather is tightly organized and coherent” (Thornbury

and Slade 2007: 27). The main point that has been presented is that speaking

has its own patterns and structures that are different from those of writing. For

example, when compared to writing spoken language uses more verbs and

clauses rather than nominalization. Furthermore, in places where embedding or

subordination might appear in a written text, speech freely adds utterances one

after another. In terms of its lexical and informational content, spoken language


is loosely packedboth to allow its audience time to process the content of

utterances and as a result of real-time processing that a speaker faces.

All in all, speaking is dynamic and is operating under conditions that

are substantially different from writing. This means that it does not always

involve using grammatically complete and written-like sentences‟ because while

written texts can be redrafted, spoken texts are results of one-shot production.

(Burns and Joyce 1997: 14) All these facts have important implications for LT,

which are to be presented in the section that follows.

2.1.2 Classroom implications: teachers’ expectations

This section explores several important implications for classroom teaching

arising from what was mentioned above. Firstly, it is crucial for teachers to

realise that spoken language is essentially different from the written one.

Teachers therefore cannot expect their students to speak in full sentences as if,

in fact, they were producing written texts. Not only is this not the way people

speak in reality but also expecting and requiring such skills from learners, would

place extremely high pressure on their speech production for no reason at all.

Such expectations might result in the learners‟ later reluctance or anxiety to


Secondly, teachers should help their students understand the

important differences between speaking and writing and instruct them in the

ways to use this knowledge effectively when speaking. For example, learners

may be less hesitant to express themselves if they are shown that speakers

string chunks of language together bit by bit without composing entire


sentences in their minds before they start to speak 4 . They may also find it

useful to learn that repairs, hesitations, repetitions and vague language are

acceptable in spoken language because without it speech production would be

made impossible. Consequently, all these aspects can be practised in class

through the use of meaningful tasks. Because as Thornbury and Slade point

out: “If this organization [meaning the organised nature of speaking] can be

described in ways that are accessible to teachers and learners, there are likely

to be practical classroom applications. (This does not mean, of course, that one

such application would simply be to „deliver‟ the description to learners without

some form of pedagogical mediation.)” (2007: 27).

My next point is that teachers involved in their own course design,

may find it useful to make their teaching more data-driven and consult further

literature dealing with detailed description of spoken language and its use of

lexico-grammatical structures. There has been a significant rise in research in

this area in recent years and books like Longman student grammar of spoken

and written English (Biber et al. 2004) or preferably Longman grammar of

spoken and written English (Biber 1999) provide quite detailed accounts of

spoken language use based on corpus research. Another book dealing with

spoken grammar in an unconventional way, which has already been mentioned,

is A Grammar of Speech (Brazil 1995).

Furthermore, teachers may also find it beneficial to make use of

spoken language corpora that are accessible on the Internet. They may even

4 Unfortunately, I could not find any evidence supporting this specific suggestion. It would be therefore valuable if this hypothesis was tested in further research. If this argument was validated through concrete evidence, it would have important implications for LT.


choose to teach corpus-based techniques in classroom for the benefit of their

learners. Conrad reports that even though “there is little empirical research into

the effectiveness of corpus-based techniques for language learning, […] there

are a variety of theoretical reasons for using them and many reports by

teachers of student interest and improvement” (Conrad 2008: 402).

Finally, it is essential for teachers to seek ways to give their learners

enough systematic training in the relevant skills, the full scope of which is to be

discussed at a later stage of this paper.

2.1.3 The role of context

As Hughes points out speaking is fundamentally transientand words are

produced “within the „co-ordinates‟ of a particular place and moment”. (Hughes

2011: 10) What follows is that speech, unlike writing, is context-dependent.

Nunan (2010) defines context as:

The linguistic and experiential situation in which a piece of language occurs. The linguistic environment refers to the words, utterances, and sentences surrounding a piece of text. The experiential environment refers to the real-world context in which the text occurs. (p. 304)

Therefore spoken texts are not created independently, regardless of

the environment, the situation or the listeners. On the contrary, speech

production takes place in a shared context between the speaker and the

listener, whereby both these participants shape the final form of a spoken text.

The more shared context there is, the easier it is for the listener to participate


in a conversation. Tannen even describes a listener as a co-author and speaker

as a co-listener (Tannen 2007).

As a result, spoken texts carry a number of specific features:

frequent use of referents like pronouns or deictic words (this, referents like pronouns or deictic words (this,

that, there) pointing to the physical context

ellipsis (deliberate omission of certain items), the meaning of (deliberate omission of certain items), the meaning of

which can be reconstructed only from the context

non-clausal stand-alone expressions such as „Yeah.‟ or „Mm.‟, expressions such as „Yeah.‟ or „Mm.‟,

whose interpretation is heavily context-dependent

(based on Thornbury and Slade 2007)

It is the context that primarily helps us reconstruct the meaning of these

utterances. By contrast, in writing, these elements are used less or avoided

completely because written texts being decontextualised, need to be as self-

explanatory as possible 5 (Burns and Joyce 1997). While ambiguities are not

desirable in written texts because there is no opportunity to provide further

explanations; the meaning of utterances that are ambiguous can be easily

negotiated in speech. Moreover, ambiguous utterances in speech are frequently

welcome because they are a source of humour, an important ingredient of daily


Furthermore, as mentioned earlier, spoken texts also contain

numerous examples of imperfect language and instances of disfluency such as

5 The decontextualised nature of writing is also one of the reasons why it is important for writers to bear their potential readers in mind and be as clear as possible. This is of course true only for texts whose purpose is informational, not texts which are read for pleasure.


hesitations, errors or repairs, which can again only be understood in the context

they are used.

Still on the subject of context, it also needs to be pointed out that

context is generally regarded as an important contributor to LA. Netten and

Planchat-Ferguson (1995) list context-defined cues along with socio-cultural

framework as one of the three principles of LA 6 . They further state that

“without these context cues and cultural norms, sensory input for acquisition

purposes may be insufficient” (1995: 35).

2.1.4 Classroom implications: teaching materials

As has been indicated, spoken texts are not normally placed out of context in

real life. The same fact should be reflected in the classroom setting. Structures

and vocabulary should be viewed in their immediate context. Longer stretches

of spoken discourse can prove extremely useful in demonstrating how different

elements combine and work together to create a successful conversational


Consequently, teachers should consider if materials and textbooks

they use reflect the features of spoken language both in their recorded and

transcribed forms. If not, they may want to consider searching for other

samples of spoken language which can be transcribed for a later systematic

training in class. Recordings are more accessible nowadays than they used to

be and it is not difficult to find desirable samples online. More high-tech

6 The other two principles they highlight are Krashen‟s concept of „comprehensible input‟ and Giles and Coupland‟s notion of integrative motivation.


teachers may even consider recording their own spoken texts. In fact, this is

something which can be done quite easily, e.g. while on holiday chatting with

your friends or asking your native-speaking colleagues to talk on a certain topic,

using specific language, etc. I personally have good experience with both using

recordings from the Internet and working in class with my own mp3 recordings

and I have seen both techniques being used with a few other teachers, too.

More than thirty years ago, Crystal and Davy (1979) complained of

the tendency of textbooks not to be real:

People in textbooks, it seems, are not allowed to tell long and unfunny jokes, to get irritable or to lose their temper, to gossip (especially about other people), to speak with their mouths full, to talk nonsense, or swear (even mildly). They do not get all mixed up while they are speaking, forget what they wanted to say, hesitate, make grammatical mistakes, argue erratically or illogically, use words vaguely, get interrupted, talk at the same time, switch speech styles, manipulate the rules of the language to suit themselves, or fail to understand. In a word, they are not real. (p. 3)

All these features are still deliberately being omitted from LT and

simplified and unauthentic materials are used. According to Burns and Joyce

(1997), inauthentic materials create a false impression of speech presenting

them with „unrealistic models of spoken interactions‟ (1997: 87).

Burns and Joyce (1997) claim that:

If the overall aim of language programs is to prepare students to use spoken language effectively in social situations, then teachers need to


present students with authentic spoken texts in the classroom. This may include the use of recordings and transcripts of authentic discourse. Teachers need to know how authentic texts differ from scripted and semi-scripted texts and how to use this knowledge to assist second language learners to develop speaking skills. (p. 85)

Furthermore, even though the authors recognise some of the

potential benefits of scripted dialogues especially at the beginning stages of

learning, they warn against their exclusive use and point out that:

) (

unrealistic view of the features of spoken language and will not be prepared for their role as participants in spoken interactions in social contexts. For students to be able participate in spoken interactions outside the classroom, the teacher will need to introduce authentic discourse gradually into the classroom. Authentic spoken texts are more difficult for students to deal with and how and when students are introduced to authentic discourse will depend on their level of language and their goals. (p. 86)

if students are restricted to scripted dialogues they will develop an

Burns and Joyce report that to eliminate the big discrepancies

between unscripted and scripted texts, some material writers choose to use

semi-scripted texts that are created by presenting several people with a

particular spoken language to be used in their interaction. In a semi-scripted

interaction, the context, purpose for interaction and specific authentic language

to be used are identified beforehand. Burns and Joyce conclude that “these

texts are a good transition between scripted and authentic texts because they

introduce students to the feature of authentic speech in a controlled way

(1997: 88).


Even though it is up to teachers when and how they introduce

authentic materials into their teaching, it is important that these are gradually

made part of LT. One of the tasks of LT is to help learners cope with real

operating conditions and real language. If authentic materials and language are

not used at all or rarely, the learners‟ picture of English language is idealistic

but not real. Therefore authentic language and materials should become an

integral part of tuition in order to help students on their way to autonomy.

Learners who are taught mainly through simplified materials may feel more

confident about their English while in class but the minute they encounter a

native speaker and experience English used under real life conditions they are

bound to fail communicatively because they are unprepared for it.

What must be made clear is that using authentic materials and/or

work with transcripts of spoken data is not to be applied exclusively or

exhaustively. Similarly to others, my experience is that adapted materials help

learners progress faster from beginning to intermediate levels. However, this

advantage should not be misused at the expense of excluding real language

from classrooms a practice that might make a language classroom a safe and

secure island for learners and teachers. It is clear that such a procedure would

not be desirable for learners in the long term. Along the same lines, Nunan

(2010) states that “learners should be fed as rich a diet of authentic data as

possible, because, ultimately, if they only encounter contrived dialogues and

listening texts, their learning task would be made more difficult” (2010: 27)

As Nunan further explains, one of the advantages of using authentic

materials in reference to context is that learners encounter target language in


the contexts where they naturally occur not where the textbook writer uses

them. In the end, this will help learners because they can experience how

language is used in relation to other closely related grammatical and discourse


To conclude, let me point out that whenever teachers feel the need

to adapt a particular lesson and include authentic materials to promote their

learners‟ learning experience, the key question to be considered is not how to

adapt the materials but how to adapt the task so that it suits the learners‟ level

of proficiency (Nunan 1989). If this is done with care, teachers need not be

afraid to include extracts from radio and television, public announcements,

telephone conversations, answering machine messages, conversations and

lectures in their classroom. Such material will make a lesson a real learning

experience and bring the content to life.

My final point on the use of materials from textbooks is that teachers

should be careful not to use textbooks as cookery books in which all ingredients

are used exactly as instructed. As Tomlinson (2008) reports, “there is evidence

that what teachers and learners actually do in the classroom is determined

principally by what the coursebook tells them to do” (2008: 143). Textbooks

have their greatest potential if they are used thoughtfully as resource books

rather than a prescriptive manual and adapted both to the learners‟ needs and

the socio-cultural context they are used in.

Similarly, if the teachers‟ aim is to teach speaking, they need to

consider if the current materials correspond with their aims and to make

necessary alterations where desirable. If they find that the books include


content that is contradictory to their aims, they should substitute relevant parts

with more suitable resources or ideally, opt for a better suited coursebook, if

they have a choice. If this is not an option, they can grab this opportunity as a

challenge to include some more authentic materials in their lessons.

2.1.5 The types of communicative exchanges

Having discussed the question of context, let us concentrate on the types of

communicative exchanges that can be identified within certain communicative

contexts. Based on a communicative situation and its purpose, two main types

of communicative exchanges can be classified: transactional and interactional.

Bygate suggests that conversations are comprised of predictable

routines. He distinguishes between information routines (called transactional by

other authors) and interactional routines (Bygate 1987). Both types differ in

their purpose and structure. Information routines consist of a number of highly

predictable language structures. Their purpose is mainly to transact goods and

services, therefore transactional (Nunan 2010). They include service encounters

such as buying a train ticket, booking a room or negotiating a loan. By contrast,

interactional routines are not product-oriented. They are social interactions and

fulfil a phatic function, i.e. they signal friendship and establish social

relationships within groups (Thornbury and Slade 2007).

Nunan (2010) illustrates different functions of both types of

exchanges in the following conversational extracts:

Extract 1:

Store attendant: Morning.


Customer: Morning. Store attendant: Nice day.

Customer: Uh-huh. Can you give me two of those?

Store attendant: Sure. Customer: Thanks.

Extract 2:

Father: Morning, Darling.

Daughter: Morning.

Father: Sleep well? Daughter: Uh-uh. The thunder woke me up. Father: Loud, wasn‟t it. And the lightning Daughter: I‟m going to finish watching that

Father: Well, don‟t have it on too loud. Jenny‟s still asleep.

What are you doing?

(p. 228)

Although it is self-evident that the purpose of the first situation is

transactional, there is an interactional element in the first part of the exchange.

Similarly, while the second extract fulfils mainly an interactional function, the

last line of the dialogue is clearly transactional (Nunan 2010). Nevertheless,

even though “many speaking situations can be a mixture of interactional and

transactional purposes” (Burns and Joyce 1997: 5), Nunan reports that

“Bygate‟s routines facilitate communication for first language speakers because

they make the interactions more predictable” (Nunan 2010: 229).

Finally, citing Brown and Yule (1983), Thornbury and Slade add that

“primarily interactional language is primarily listener-oriented, whereas primarily

transactional language is primarily message-oriented” (Thornbury and Slade

2007: 20). Viewed from this perspective, one can come to a realization that

listener-oriented interactions will tend to be freer in terms of their structure.


This is mainly because interactional conversational exchanges can easily deviate

from their primary focus reflecting the listener‟s personal involvement. On the

other hand, message-oriented conversations will be more clearly structured, as

evidence shows, pursuing their ultimate objective to deliver a message.

2.1.6 Classroom implications: communicative tasks use

The fact that interactions can be predictable has important implications for

teachers. They can help their students prepare for both types of conversational

exchanges: transactional and interactional. Since these have different features

and a structure, they require different skills to be trained.

On one hand, service encounters are relatively easy to deal with for

their easily predictable structures (Brown and Yule 1983) and students will

strongly benefit from practising various kinds of transactional interactions

because these will help them prepare themselves for real-life situations.

Moreover, Nunan (2010) stresses the fact that learning prefabricated

conversational patterns enables learners to „outperform‟ their competence. This

means that learners can become relatively confident in many situations even

with rather limited language means at their disposal.

On the other hand, interactional exchanges, in comparison with

transactional ones, will probably bring more life and enjoyment in the classroom

thanks to their free nature and eventual inclusion of sharing personal

experiences. Moreover, the latter point is believed to be a motivating factor

facilitating LA based on the evidence from peer-interaction studies. Willis (1996)

even lists sharing personal experiences as one of task types to be used in the


classroom. Furthermore, as early as 1961, Billows stressed the importance of

„personal speech‟ in teaching (Thornbury and Slade 2010). Thornbury and Slade

also support the use of personalised talk by reporting that it has even become

an integral part of one of the methodological approaches, namely Community

Language Learning developed by Curran in 1976.

Thornbury and Slade complain of the methodologists‟ preference of

transactional LT and their choice to neglect free open discussions in language

textbooks. They remark that many classroom activities, such as role plays, tend

to be too prescriptive (2010: 266). Consequently, classroom exchanges

whether interactional or transactionalhave a tendency to turn into

transactional turns leaving not much room for free interaction. By the same

token, if instructions and workplans are too detailed, learners numbly follow

their patterns in pre-given sequences focusing on task completion without the

need to employ their pragmatic and discourse skills. This is because they are

simply not required to think about a discourse as a whole and produce it


An example of such an activity is Activity 1 (see Appendix C), a task

that is also mentioned by Thornbury and Slade (2007), in which learners act out

a phone call following detailed guidelines. In each turn they are asked to

respond in a certain way following the diagram. My personal experience with

this task is that for many learners the detailed guidelines are rather counter-

productive. With these types of tasks it frequently happens that learners want

to add some extra information and by doing so, their conversation no longer fits

the pattern and they do not know how to proceed. This causes confusion and


sense of failure in the task completion. Other learners who, on the other hand,

follow the given pattern to the letter usually end up quoting parts of the given

phrases or sentences without changing them too much.

It is true that at beginning stages strictly guided tasks may help

learners focus on how to say things rather than think about what to say next

and thus extend their communicative potential at least for the course of the

task. However, such training will not equip them with the skills that are

necessary to engage in similar interactions on their own in the long term.

Therefore, since the number of skills that learners practise in this way

is highly limited and such an activity has a tendency to turn into a drill exercise

rather than a conversation, freer tasks might be more appropriate allowing

learners to apply a broader range of skills. In terms of communicative practice a

better way of doing the same thing, that is practising telephone calls, is Activity

2 and even better Activity 3 (Appendix C), since learners are given more

leeway. In activity 2, learners A and B receive a different set of information and

act out their roles accordingly. Even though they are asked to convey a

particular message they are not told exactly in which part of the conversation to

do it. Nevertheless, once again the instructions are given exactly in the order in

which the conversation is likely to proceed based on the instructions the other

learner has.

By contrast, Activity 3 is a bit more genuine because it describes the

situation the learner is in and gives learners only general instructions such as

answer the telephone and take a message”. This is the situation that

resembles real life most because in reality learners will hardly ever know who is


going to call and how they are supposed to react. Therefore using this type of

transactional task in class is likely to prepare the learners for real-life situations

best. Apart from this, learners are likely to enjoy the task more because using

cards instead of following instructions from a book is more interactive as well as


In comparison with transactional, interactional tasks can be much

freer, such as Activities 4 and 5. Activity 4 provides picture prompts and

questions which give learners a general idea of what they might talk about.

Nevertheless, the way they handle the task is entirely at their discretion. Apart

from pictures representing mainly popular American series, the activity also

offers culture-specific prompts which are likely to instigate more complex

interaction. Activity 5 is a success with learners because it is them who decide

how long each topic will be discussed and what questions will be asked. If they

are asked to pose questions and respond to them as quickly as possible, the

communicative exchanges can become very fast, highly competitive and

learners usually have a good laugh at it.

Since many communication exchanges require both transactional and

interactional language, a good example of how to prepare learners for that are

role-play cards which are cut up and distributed in pairs. Learners take turns in

picking up individual cards and act out the situations (Activity 6).

On the issue of interactional tasks, Thornbury and Slade indicate that

freer tasks such as having a conversationare generally avoided in present

classroom teaching. They illustrate it on an example of task-based language

teaching (TBLT). They say that even though proponents of TBLT advocate


communicative environment creation in class and put their emphasis on

pragmatic language processing and meaningful language use which resembles

“the way language is used in the real world” (Ellis cited in Thornbury and Slade

2010: 267), conversation in itself does not seem to qualify as a task type in

task-based materials because it is not a clearly structured and goal-oriented

activity. They further explain that some scholars, such as Skehan, are sceptical

as to the value of conversation. In Skehan‟s view “the elliptical and jointly

constructed nature of conversation is not conducive to the production of well

formed sentences, and speakers are able to „bypas syntax‟ a great deal of the

time” (Skehan cited in Thornbury and Slade 2010: 269). However, based on the

findings in 2001, Nakahama et al. suggest that “conversation should be studied

in much more detail as a potential source of rich learning opportunities” (in

Thornbury and Slade 2010: 270), a conclusion which is consistent with

Thornbury and Slade‟s line of argument.

As a result of a long-lasting tradition prioritizing transactional

activities over interactional ones, one of the challenges of language teachers

when teaching interactional language exchanges may be the shortage of

relevant resources. In the case of English, innovative coursebooks integrating

communicative activities or at least offering materials that can be adapted

are fairly rare but can be found nowadays. However, as far as LT in general is

concerned, the situation is pitiful. This is especially true with textbooks and

resources designed in countries where the tradition of form-focused LT has

been particularly strong: the Czech Republic, Austria and Germany being

illustrative examples. Thus, German teachers, for example, have to resort to


designing their own materials when teaching speaking. Unfortunately, searching

for communicative activities in German coursebooks, resembles looking for a

needle in a haystack 7 .

In summary, both transactional and interactional activities are highly

useful and should be included in LT in a sufficient measure to provide plentiful

opportunities for learners to practise using both types of communication

exchanges. In many cases, teachers will need to prepare extra materials to

cover both types of communication adequately in their lessons.

2.1.7 Characteristics of conversation

Coming back to the issue of free interaction practice, one cannot but notice the

fact that conversation is one of the most common types of speech production

that people are involved in on a daily basis (Thornbury and Slade 2007: 5). For

this reason, it is important for us to look at it briefly as a genre of its own.

Thornbury and Slade define conversation as follows:

Conversation is the informal, interactive talk between two or more people, which happens in real time, is spontaneous, has a largely, interpersonal function, and in which participants share symmetrical rights. (2007: 25)

7 This statement is based on my five-year experience teaching German and using materials available in the Czech market by German publishers, in particular by Hueber, Klett and others. Having conducted a short empirical research of coursebooks by Hueber for the purpose of this thesis, I found only a few activities which could be identified as communicative. The general outline of all Hueber textbooks shows a strong accuracy-focused pattern, placing grammar and vocabulary in the centre of LT.


Conversation, involving a two-way interaction between people,

requires from its participants a broad range of skills. Participants need to know

how to interact and manage talk. They need to be familiar with the rules of

turn-taking, that is they need to know when and how to interrupt, take the floor

and how to hold it, how to change the topic, how to signal they wish to speak

and how to yield the turn. Furthermore, it is important for them to know how to

signal interest and the fact that you are listening or even how to avoid long

silences (Thornbury 2005).

Quite importantly, all of these skills need to be employed readily at a

moment‟s notice usually several of them at a time. For this purpose, speakers

need to be equipped with a number of discourse markers to signal the others

what their intentions are. Speakers also cannot do without paralinguistic cues,

such as various body movements, gestures, appropriate use of eye contact,

intakes of breath and so forth.

One of the concepts that is very important from a methodological

point of view for many types of conversational exchanges, is the concept of

adjacency pairs. Thornbury and Slade emphasise that “(…) the basic unit of

interaction is the adjacency pair” (Thornbury and Slade 2007: 114). They

further explain that “an adjacency pair is composed of two turns produced by

different speakers which are placed adjacently and where the second utterance

is identified as related to the first” (2007: 115). Adjacency pairs typically include

stereotypical exchanges such as question/answer; complaint/denial;

offer/accept; request/grant; compliment/rejection; challenge/rejection, and


instruct/receipt. A successful conversation typically includes a number of such

exchanges with one speaker initiating the move and the other one responding.

Another fact that is stressed by some scholars is that the success of a

conversation depends on how cooperative both speakers are (Tannen 2007).

Typically, the more cooperation there is, the fewer overlaps occur. The same

applies for turn-taking. Smooth transitions between turns usually happen with

speakers that collaborate (Burns and Joyce 1997). As a result, conversation

always needs to be viewed as an act of multiple parties where each individual is

responsible for the potential success or failure of a communication.

Consequently, it can be said that conversation is greatly shaped by

individual personalities of all participating members and their styles of

communication. This means that even under ideal conditions, when all speakers

converse in their L1, not all conversations are successful. If for whatever reason

one of the speakers is uncooperative, this will be reflected in the conversation.

Similarly, Hughes 2006 refers to evidence that communicative success of L2

speakers in L2 environment depends on the attitude of native speaker towards

non-native speaker. If this is negative, the communication is more likely to fail.

By analogy, conversations between speakers who have different levels of

knowledge may still be highly successful if all the participating members

cooperate; for instance, if native or highly advanced speakers are willing to

provide assistance to those whose level of language proficiency is lower.

In summary, to actively participate in a conversation requires a broad

range of skills. Speakers need to know how to interact and manage the talk and

need to be familiar with the rules of turn-taking. Apart from these, speakers


need to know how to effectively realise initiating and responding moves

through the use of the so-called adjacency pairs. Nevertheless, however high

their level of proficiency might be, the success of a conversation still depends

on all members taking part in a particular conversational exchange.

2.1.8 Classroom implications: challenges of conversational classes

To be able to engage oneself in a conversation using all the above-mentioned

techniques requires a great deal of skill. To become a truly effective participant

of a conversation means to be able to interact with others spontaneously under

many different and unexpected conditions and about a broad range of topics.

In this respect, purely conversational classes can be a challenge both

for teachers and learners. As a learner, it is challenging to become an effective

interlocutor in a foreign language especially if one does not possess these skills

in one‟s L1 in the first place. By analogy, it is similarly challenging (if not

impossible) to become a good teacher of conversational classes if one is not

aware of all the skills that speakers need and/or if one cannot utilise them

effectively oneself. As mentioned earlier, one‟s personality also plays an

important role here because it fundamentally influences the nature of a


Along the same lines, one thing I have observed is that it is essential

for learners to have teachers who they can learn conversational patterns from

on a daily basis by observing the ways their teacher interacts in class. For

teachers to be able to serve as good models as far as conversation is concerned

is essential for several reasons.


Firstly, it is generally believed that languages are learned partly

through imitation (Lightbown and Spada 2006). Secondly, evidence shows that

teachers are a source of input for learners, even more so in an EFL context

where the language lessons might be the only opportunity for some learners to

listen to English being used. Thirdly, the teachersrole in the exposure of

learners to language also should not be underestimated because it is a proven

fact that some aspects of language are acquired subconsciously. Lastly, being

able to experience various conversational patterns in their natural context

enables students to gain unique language experience that is likely to have a

positive effect on their LA. Depriving students of these opportunities, on the

other hand, is a fundamental pedagogical failure. By the same token, this is

what Lightbown and Spada complain about when saying that “in many foreign

classes, teachers switch to their students‟ first language for discipline or

classroom management, thus depriving learners of opportunities to experience

uses of the language in real communication” (2006: 32).

From my experience, another fact that teachers face when teaching

conversation is that learners often do not realise the complexity of skills that is

required but attribute their eventual failure to inadequate knowledge of

language. Diligent students often try to learn as many new words as they can in

hope that their conversational skills will improve as their vocabularies grow.

However, the result of such striving can only be disappointing because as can

be inferred, no knowledge of lexis or grammar can substitute for inadequate

interaction skills. Undoubtedly, it is a teacher‟s role to illustrate the complexity

of skills that are required and help their students acquire them. Secondly, it is


also a good idea to help learners understand one of the fundamental features

of conversation discussed in the previous section, which is that conversation is

a two-way interaction. Learners therefore cannot attribute eventual failure of

conversation to their linguistic incapacities.

All in all, teaching conversation amounts to teaching interaction and

everything that it entails. Some of the skills that are necessary for one to be

able to converse have been touched upon. The whole spectrum of knowledge

and skills that learners need in order to become competent speakers will be

focused on in the following part of this chapter. For, as one might infer from

what has been mentioned so far, the ability to speak a language involves much

more than mere linguistic knowledge. The range of knowledge and skills that

are necessary are called communicative competence, the notion which is

presented in the next section.

To sum up, this section aimed to introduce some of the main

characteristics of spoken language that make speaking and its production

significantly different when compared to writing. It also sought to link the

individual features of speaking to their implications for classroom teaching.

On a final note, it needs to be pointed out that for teachers, it is

especially important to be aware of the above-mentioned differences between

speaking and writing. If teachers take them into consideration, they will more

likely be able to create lessons that will truly help learners acquire skills that are

needed for an effective oral communication.


2.2 Communicative competence

2.2.1 Historical background

The notion of communicative competence was introduced by Hymes in reaction

to a restricted Chomskian concept of linguistic competence. In 1965, Noam

Chomsky made a distinction between competence, an ideal picture of speaker-

listener‟s knowledge of his language, and performance, the actual use of

language in concrete situations‟ (Chomsky 1965: 4). Chomsky further explained

that as a record of natural speech, performance is imperfect carrying numerous

false starts, deviations from rules, etc.

In contrast, Hymes (as cited in Magnan 2007: 350) defines

communicative competence as the ability „to participate in [the child‟s] society

as not only a speaking, but also a communicating member‟. Hymes‟s

understanding of competence is much broader than to comprise only a

linguistic perspective. Trosborg notes that Hymes expands the definition to

encompass “all rule-systems underlying language use, and thus accords a

central role to sociocultural factors(Trosborg 1986: 9). On the same page, she

further adds that the work of Hymes “exemplifies the shift away from the study

of language as a system in isolation towards the study of language as


Using the notion as a broader term similarly to Hymes, Allwright even

implies that some areas of linguistic competence are essentially irrelevant to

communicative competence, but that, in general, linguistic competence is a part

of communicative competence(Allwright 1979: 168). Looking at the diagram

provided by Allwright, it needs to be observed that should the diagram be


representative of the scope of both competences, Communicative Competence,

from what we know nowadays, would need to be portrayed as much a bigger

circle in comparison with Linguistic Competence (see Diagram 1 below

portraying the relationship between communicative and linguistic competence):

between communicative and linguistic competence): Diagram 1 (Allwright 1979: 168) Canale and Swain extend the

Diagram 1 (Allwright 1979: 168)

Canale and Swain extend the model of communicative competence to

comprise four interrelated areas of competence: linguistic competence,

sociolinguistic competence, discourse competence and strategic competence.

Let us take a brief look at what these four dimensions represent.

Linguistic competence refers to the mastery of the language code‟.

Sociolinguistic competence focuses on the sociocultural rules of use, i.e. the

system of rules which determines the appropriateness of a given utterance in a

given social context‟. Discourse competence represents „the appropriateness of

utterances to their linguistic contexts‟, i.e. „the knowledge how to combine

sentences into unified spoken or written texts of various types‟ (Trosborg 1986:

911). In Swains words, strategic competence can be described as:

the mastery of the communication strategies that may be called into action either to enhance the effectiveness of communication or to compensate for breakdowns in communication due to limiting factors in actual communication or to insufficient competence in one or more of


the other components of communicative competence (Swain cited by Scarcella and Oxford 1992: 72).

In other words, strategic competence refers both to tools that help

negotiate the meaning and compensatory tools that speakers employ to

overcome real-time processing constraints. This is a very interesting definition

as we shall see in the next section of this chapter.

So far, the notion of communicative competence has been discussed

mainly from the perspective of Canale and Swain‟s four-dimensional framework.

There have been several more attempts to describe language knowledge and

skills that competent speakers have at their disposal. For example, a more

recent model by Bachman expands the concept of communicative language

ability to include several broad areas: organizational competence including

grammatical and textual competence, pragmatic competence including

illocutionary and sociolinguistic competence, strategic competence and

psychophysiological mechanisms 8 (Magnan 2007). Other models also draw

heavily on Canale and Swain‟s model from 1980.

All in all, the crucial point to be made here is that through the use of

CLT it has become widely understood that communicative competence should

be the primary goal of LT (Savignon 1997). The eventual model that is chosen

to describe this concept is secondary to this realization.

8 This concept was expanded further by Littlemore and Low (2006), who stressed metaphorical competence to be a part of communicative language ability. Celce-Murcia et al. (1995) moved discourse competence to the centre of the model, stating that it is shaped by sociocultural, linguistic and actional competences. Nevertheless, their five-dimensional model also does not abandon the concept of strategic competence proposed earlier. (Magnan 2007).


To conclude, since all components of communicative competence are

interrelated, as Trosborg (1986) mentions; it is crucial for us to consider all the

above-mentioned dimensions. As can be inferred, teaching any of the

competences as well as any of the four skills means to teach the others, too

because LT is highly interconnected. Teaching speaking skills is no exception

here (Hughes 2011). Along the same lines, Bygate argues that for teachers the

central problem in conceptualising the teaching and learning of speaking may

be the fact that “it can be hard to disentangle the different dimensions of

foreign language speaking” (Bygate 2009). In the next section, the focus will

therefore be shifted to the individual aspects that learners need to incorporate

in order to achieve communicative competence.

2.2.2 Individual components of communicative competence

This section aims to discuss the individual components comprising

communicative competence. Although the concept of communicative

competence can be viewed both from the perspective of spoken and written

forms of a language, the angle that is taken here is that of the spoken language


Furthermore, even though Canale and Swain point out that there are

two basic dimensions to communicative competence: knowledge and skill

(Trosborg: 1986), this chapter discusses mainly the former, i.e. the knowledge

that one needs to possess to become a competent speaker. However, it should

not be forgotten, that attaining communicative competence would not be

possible without a great deal of skill.


The concept of „skillrepresents the readiness with which one is able

to combine this theoretical knowledge in order to conceptualise, formulate,

articulate and monitor their thoughts with a certain degree of automaticity

(Thornbury 2005). Thornbury also stresses the fact that speaking is “like any

other skill, such as driving or playing a musical instrument: the more practice

you get, the more likely it is you will be able to chunk small units into larger

ones” and achieve fluency (2005: 6). Knowing that both concepts of knowledge

and skill are interrelated, let us focus now on the description of the former so

that some forms of practising the latter can be presented in the next chapter.

It has been already pointed out that communicative competence

consists of four dimensions: linguistic, sociolinguistic, discourse and strategic.

Linguistic competence that is to be discussed here encompasses the knowledge

of grammar, lexis, syntax, phonetics, etc. The following sections focus on some

of the above-mentioned areas and discuss the aspects which are important for

speaking skills teaching.

The grammar of spoken language is distinguished from written

grammar by several important features. Thornbury lists them in a clearly

arranged table (see Table 1 at the top of the next page).


Table 1 (Thornbury 2005: 21) The majority of the above-mentioned aspects, such as clause addition

Table 1 (Thornbury 2005: 21)

The majority of the above-mentioned aspects, such as clause addition (add-on

strategy or layering of phrases/clauses), ellipsis and hesitations, etc. have

already been touched upon in the previous sections of this thesis. The

remaining features result from the inherent nature of speaking in a similar way

to what has been described. For example, time constraints can explain for the

tolerance of vagueness and a two-way character of interaction in conversation

for the frequent use of question tags.

Other important facts which might be useful for teachers whose

learners aim to attain the spoken rather than the written form of a language

are summed up by Thornbury (2005) as follows:

present forms outnumber past tense forms by 2:1.of a language are summed up by Thornbury (2005) as follows: simple forms outnumber progressive and

simple forms outnumber progressive and perfect forms by overform of a language are summed up by Thornbury (2005) as follows: present forms outnumber past



the past perfect and present perfect continuous are rare.passive verbs account for only 2% of all finite verb forms in speech. will, would,

passive verbs account for only 2% of all finite verb forms in speech.the past perfect and present perfect continuous are rare. will, would, and can are extremely common

will, would, and can are extremely common in speech. (p. 22) and can are extremely common in speech. (p. 22)

By including this list, I do not mean to suggest that progressive

forms, past and present perfect continuous and passive voice should be

excluded from tuition. However, the time that might be wasted to help learners

attain an active command of these phenomena, might be better spent

practising other aspects of communicative competence that would actually

make a difference in terms of learners‟ practical communicative abilities, e.g.

sociolinguistic, discourse and strategic competences.

Having mentioned grammar, it needs to be pointed out that some

important discoveries have been made, which unfortunately have not yet been

properly reflected in LT syllabuses. In 1973, Brown and his colleagues found

that FLA of grammatical morphemes in English occurs in similar sequences with

all children (Lightbown and Spada 2006). To put it in a nutshell, morphemes

that are first on the list to be acquired are for example present progressive -ing

(Mommy running), plural -s and irregular past forms (Baby went), while those

that are acquired at later stages are third person singular simple present s

(She runs) and auxiliary be (He is coming). (2006: 3)

Similar acquisition order has been found with second language

learners as Nunan (2008) reports. On balance, it needs to be said, however,

that some studies contradict the clear-cut pattern of these findings, albeit a

number of similarities can be observed. Nevertheless, research has also


confirmed that some aspects like the formation of questions or sentence word

order are acquired at later stages of LA (Lightbown and Spada 2006).

According to Cook (2009), the current version of this idea is

Processability Theory. This theory, proposed by Pienemann, claims that learners

start their spoken production by producing single content words such as water,

flower, etc. Then they learn how to combine words and put them in the right

order within one sentence. After that they learn how to move parts of the

sentence around to ask questions Does Jane like flowers? Finally, they learn to

use the right order in complex sentences such as reported speech I asked if

Jane liked flowers.

In spite of the fact that many EFL and ESL textbooks do not reflect

these findings, teachers should be careful about introducing these

morphologico-syntactical aspects too early into tuition or to persevere in their

teaching if learners are not ready to incorporate them into their L2 language

systems (see Dulay 1982).

Another significant part of linguistic knowledge to be discussed here

is the knowledge of lexis. Lexis does not represent only individual words, word

families and their usage in context but it also includes formulaic language such

as collocations, phrasal verbs, idioms and sayings, sentence frames such as

would you like a


social formula like have a nice day, discourse markers,

e.g. if you ask me etc. (Thornbury 2005).

In reference to the range of vocabulary that is needed for spoken

interaction Thornbury and Slade (2007) quote Nation saying that “from the


small amount of evidence available, it seems that about half the words needed

to understand written English are needed to understand spoken English” (2007:

42). Thornbury (2005) adds that “according to some estimates, a vocabulary of

just 2,500 words covers nearly 95% of spoken text (compared to 80% of

written text)” (2005: 23). This is good news, since a similar percentage (90%)

is generally quoted as necessary for one to be able to make sense of any text,

whether written or spoken (e.g Lightbown and Spada 2006). What follows is

that the knowledge of roughly 2,500 words should be a sufficient range to

provide learners with an operational command of English lexis for spoken


It is beyond the scope of this thesis to discuss various vocabulary

learning strategies; nevertheless, let me add a marginal note that Oxford

(2011) lists a number of those that are considered positive according to

theoretical or empirical studies.

My last point in relation to the acquisition of lexis is that, FLA

research shows that vocabulary growth is mainly dependent on how widely

children read, as Lightbown and Spada point out. Here they refer to the studies

of Nagy, Herman, and Anderson 1985 and Dee Gardner 2004 (Lightbown and

Spada 2006). To support vocabulary growth of learners, teachers should

therefore encourage learners to read extensively. Designing collective reading

plans may be a good idea as well.

Sociocultural knowledge or to use Canale and Swain‟s terminology,

the knowledge that forms sociolinguistic competence is composed of „social


values and the norms of behaviour in a given society, including the way these

values and norms are realised through language‟ (Thornbury 2005: 12). This

kind of knowledge can be both extralinguistic and linguistic. Thornbury

illustrates this fact with the following example: “knowing whether people in a

given culture shake hands on meeting, or embrace, or bow, is extralinguistic;

knowing what they say when they greet each other is clearly linguistic” (p. 12).

Thornbury and Slade (2007) provide a humorous yet apt illustration: “it is a

sociolinguistic error to greet someone by saying good night instead of good

evening, but a sociocultural one to individually greet all the occupants of a lift”

(2007: 228).

Sociocultural knowledge is a type of knowledge that is typically

acquired within one‟s own culture. Unfortunately, the bigger the gap between

L1 and L2 cultures, the more probable it is for the transfer of sociocultural

norms from L1 to have a negative rather than positive effect on sociocultural

competence acquisition (Thornbury and Slade 2007: Chapter 7).

Sociocultural knowledge should not therefore be underestimated.

According to research evidence, sociocultural rules belong to that part of

language knowledge that is best learned through instruction (Thornbury and

Slade 2007). If left to work out the rules for themselves, learners will find it

hard to counter the effects of negative transfer from their L1 relying only on LA

processes. With respect to the evidence, nowadays it is widely accepted among

scholars that sociocultural competence should be mediated in language

classrooms and various integrative approaches to teach culture along with

language have emerged since the 1990s. (See Savignon 2008) The question


remains, however, to what extent teachers are successful in mediating this


On the same subject, the fact that standard textbooks often choose

materials to suit a broad range of cultures does not make it easier for teachers

to educate their learners in the sociocultural area. As a result, the range of

topics that are traditionally covered in textbooks is rather limited and generic

across publishers. To reconcile this division between the contents of textbooks

and learners‟ needs, EFL teachers need to search for suitable materials and

incorporate culture-specific topics to be contrasted in their lessons with respect

to the sociocultural background of their learners.

Another important aspect of sociocultural competence is what

Thornbury (2005) calls intercultural competence, i.e. the ability to manage

cross-cultural encounters irrespective of the culture of the language being

used”. As Thornbury explains “knowing how to ask How do you do that here?

may be more useful than a list of „dos and don‟ts‟” (2005: 32).

Finally, as suggested above, sociocultural competence does not

merely represent the knowledge of values and the norms of behaviour in a

society, but also the way these are realised through language. The latter aspect

is to be pointed out here because negative transfer of discoursal and

sociolinguistic features from the learners‟ L1 into L2 “may have much more

serious consequences than errors at the level of syntax or pronunciation,

because conversational competence is closely related to the presentation of

self, that is, communicating an image of ourselves to others” (Thomas 1983).

Thus violating the language use in terms of politeness will have more serious


impact than any other linguistic error. This is especially true if the speaker has

reached a higher level of communicative language competence. Unfortunately,

such errors will be understood not as a bad language use but rather as irritating

personal characteristics of the speaker.

Therefore it is absolutely essential to instruct learners in pragmatic

aspects of language, such as speech acts (see Austin 1962), frequently also

called functions. In general, speech acts cover a broad range of specific

discourse moves such as ways of giving advice, suggestions etc. Examples of

these might be: you may want to consider




, you ought to


what I suggest

Because speech acts theory interrelates several areas of language, it

is not easy to place speech acts in one of the four dimensions of communicative

competence. Therefore they can be seen as part of sociocultural, discourse or

even strategic competence. Nevertheless, the important thing is that speech

acts or functions will enable learners to “choose the words which most suitably

realize their intention, and this does not always entail the most closely related

form” (Cook 1989: 41). Since the choice of words may often not be intuitive or

cannot be transferred from L1, there is a good case for the explicit teaching of

speech acts as Thornbury (2005) advocates.

The discussion of speech acts leads us to the third area of

communicative competence, i.e. discourse competence. Even though not

named specifically, this dimension has already been touched upon when

discussing the characteristics of conversation (see section 2.1.7). To sum up


briefly what exactly falls under this category, discourse knowledge involves

knowing how speakers realise their turns and signal their intentions through the

use of discourse markers. Discourse competence also enables speakers to be

coherent and use register appropriately.

The former, coherence and cohesion, are partly realised in a manner

that is different in speech from that, which is used in writing. One of the

aspects that make spoken texts hang together is lexis and its repetition.

Although written texts also make use of semantic cohesion, the use of lexical

chains unlike simple repetition is preferred (Cook 1989). In addition, spoken

texts also make use of reduplication (e.g. She inspected, her passenger, this

little old lady), retrospective labelling (e.g. Well, thats what she said!), etc.

(Brazil 1995). Syntactically, the texts do not contain complex embedded

sentences and subordinations as do written texts. Utterances are linked by the

so-called add-on strategy 9 .

Which of these aspects should be taught is a question. In reference to a

part of discourse competence, in particular the knowledge how to manage

turns, Thornbury (2005) notes that “since this is a universal feature of spoken

interaction, it is not something learners need to be taught. They simply need to

know how these turn-management moves are realized in the second language,

through the use, primarily, of discourse markers” (2005: 33). On the other

hand, it is clear that certain aspects such as register and the correct usage of

discourse markers will more easily be acquired through formal teaching. One

9 The way how utterances are strung together in spoken discourse has been discussed in section 2.1.1 of this thesis.


fact should not be neglected when deciding for or against treating some of

these features. This is that there is an extensive literature on the potential for

misunderstandings due to negative transfer, specially of discoursal and

sociolinguistic features of the learners‟ L1”, as Thornbury and Slade put it

(2007: 227).

Before discussing strategic competence, the notion of

„communication‟ strategies needs to be introduced first.

Communication strategies, a term coined by Selinker in 1972 (Elllis

1985: 180), are an important part of TEFL. A number of scholars note that their

use enables speakers (or language learners in the context of L2 acquisition) to

compensate for language deficiencies. This tradition of seeing communicative

strategies in terms of failure, that is as a compensatory means used to repair

broken language, has been particularly strong as Cook (2009) points out.

Communication strategies can be viewed from another perspective as

a positive tool that helps listeners to „negotiate meaning‟. Hesitations, word

repetition and repairs are therefore not perceived as communication

breakdowns but as an effective tool enabling interlocutors deal with challenging

conditions which oral exchanges bring. This tool also helps the interlocutors to

understand the meaning of utterances that might otherwise be misinterpreted

or not understood at all. Moreover, it also helps them express thoughts they

would otherwise not be able to convey.

Canale and Swain suggest that communication strategies are to be

seen as a part of communicative competence. Instead of the term


communication strategies they identify the concept as „strategic competence‟

(Trosborg 1986). This term is certainly more precise because it also covers the

concept of „production strategies‟, which according to Thornbury and Slade

(2007) are often used in an attempt to use a linguistic system „efficiently and

clearly, with a minimum of effort‟ (2007: 220). By contrast, communication

strategies are used in the search for missing linguistic knowledge. Nevertheless,

Thornbury and Slade further admit that both concepts overlap and it may often

be difficult to determine what motivates the use of each.

Still on the subject of the divided perception of strategic competence,

it is to be noted that Swain‟s definition of strategic competence covers both

sides of the coin because she says that this term refers to:

the mastery of the communication strategies that may be called into action either to enhance the effectiveness of communication or to compensate for breakdowns in communication due to limiting factors in actual communication or to insufficient competence in one or more of the other components of communicative competence (Swain cited by Scarcella and Oxford 1992: 72).

In other words, strategic competence refers both to tools that help

negotiate the meaning and the compensatory tools that speakers employ to

overcome real-time processing constraints. Niżegorodcew (2007) summarises

the concept by saying that: “Positive strategic behaviour leads to the

achievement of the communicative goal while the negative strategic behaviour

means avoiding difficulties or transferring the responsibility to the interlocutor”

(2007: 34).


In terms of whether to teach strategic competence, scholars‟ answers

are not completely unified. For example, Cook (2009) states that “teachers may

not need to specifically teach communication strategies” but “simply encourage

students to make use of those they already prefer” (2009: 145). She reaches

this conclusion on the basis of Poulisse‟s study (1990) carried out at the

University of Nijmegen, in which learners demonstrated that their L2 strategies

largely reflected the strategies which they used in their first language. Cook

therefore infers that strategic competence teaching is not necessary because

learners can simply transfer the strategies they already use in their L1.

By contrast, Niżegorodcew (2009) recommends strategic competence

training, even though in reference to Solarczyk she admits that there are still

few studies providing evidence for significant results. However, she makes a

reference to her own research stating that “positive strategic behaviour can

slightly increase low proficiency L2 learners‟ competence”. Nevertheless, she

also points out that “the most significant factor influencing communicative

competence of the learners is their linguistic competence, that is, L2 proficiency

level” (2009: 34).

In consistence with Niżegorodcew‟s view, Scarcella and Oxford (1992)

also define the strategic competence as a tool allowing effective speakers “to

stretch their ability to communicate effectively in their new language” (1992:


My personal view of this issue is that strategies improving

communication abilities can help all speakers a) to be more effective and clear

in communicating their message and b) to easily overcome time-processing


demands. This is true for all kinds of verbal expression, whether speaking L2 or

one‟s own L1, a fact which I have observed as a teacher and a language learner

and also as an occasional participant in sessions of Toastmasters International,

a non-profit educational organization teaching public speaking skills.

The range of skills that is needed for successful communication is

really broad. To summarise let me present a specific list of features that are

particularly useful for learners, as recommended by Richards (1990) for

teaching conversation:

how to use conversation for both transactional and interactional purposesas recommended by Richards (1990) for teaching conversation: how to produce both short and long turns

how to produce both short and long turns in conversationfor both transactional and interactional purposes strategies for managing turn-taking in conversation,

strategies for managing turn-taking in conversation, including taking a turn, holding a turn, and relinquishing a turnhow to produce both short and long turns in conversation strategies for opening and closing conversations

strategies for opening and closing conversationstaking a turn, holding a turn, and relinquishing a turn how to initiate and respond to

how to initiate and respond to talk on a broad range of topics, and how to develop and maintain talk on these topicsa turn strategies for opening and closing conversations how to use both a casual style of

how to use both a casual style of speaking and a neutral or more formal styletopics, and how to develop and maintain talk on these topics how to use conversation in

how to use conversation in different social settings and for different kinds of social encounters, such as on the telephone and in informal and formal social gatheringscasual style of speaking and a neutral or more formal style strategies for repairing trouble spots

strategies for repairing trouble spots in conversation, including communication breakdowns and comprehension problemsthe telephone and in informal and formal social gatherings how to maintain fluency in conversation through

how to maintain fluency in conversation through avoiding excessive pausing, breakdowns, and errors of grammar or pronunciationcommunication breakdowns and comprehension problems how to produce talk in a conversational mode, using a

how to produce talk in a conversational mode, using a conversational register and syntaxpausing, breakdowns, and errors of grammar or pronunciation how to use conversational fillers and small talk

how to use conversational fillers and small talkin a conversational mode, using a conversational register and syntax how to use conversational routines (p.

how to use conversational routinesmode, using a conversational register and syntax how to use conversational fillers and small talk (p.

(p. 7980)


To sum up, speaking and successful communication require

knowledge of linguistic, sociolinguistic, discourse and strategic features. Even

though to some it may seem that linguistic dimension such as knowledge of

grammar structures and vocabulary are the major aspects of communicative

competence, research shows that deficiencies in other areas such as discourse

and sociocultural competence may have more significant impact on

communication than incorrect grammar use. On the other hand, it needs to be

pointed out that overall it is one‟s language proficiency that has a significant

impact on the character of one‟s performance. Thus, even though

communication strategies can help learners bridge certain gaps in their

knowledge, they cannot substitute for one‟s overall low language proficiency.

Therefore effective speaking skills teaching involves systematic development of

all dimensions of communicative competence assisting learners in achieving

higher levels of proficiency.

Taking everything into account, one needs to remember that apart

from requiring a certain knowledge base, speaking is primarily a skill. This skill,

called fluency, will be presented in the next chapter.


3 Practical part

This part focuses on two important objectives of speaking skills lessons: fluency

and accuracy. In addition, it also touches upon error correction briefly and

emphasises the importance of pair work in LT.

Apart from this, the aim of this chapter is to present a range of

techniques which can be recommended for speaking skills practice. Most of the

examples of the individual techniques which are included here were chosen

especially because I have observed in my lessons that they had a positive effect

on the learners‟ spoken production. In general, they helped learners progress in

their communicative competence increasing their fluency and/or accuracy and

learners enjoyed them. The majority of the tasks presented here are fluency

focused because in EFL textbooks, fluency-practice activities usually receive

substantially less attention than other types of activities. However, it needs to

be pointed out that the range of task types presented here is by no means


3.1 Individual aspects of teaching speaking skills

3.1.1 Fluency and accuracy

Fluency/accuracy dichotomy is one of the concepts which usually come to mind

first when speaking of teaching speaking skills. According to Segalowitz (2003:

384), the term „fluency‟ is “an ability in the second language to produce or

comprehend utterances smoothly, rapidly, and accurately”. This definition is

interesting because it clearly shows that both concepts, fluency and accuracy,


are closely knit together. Technically speaking, the term fluency is a hypernym

because to be fluent means not only to „produce utterances smoothly and

rapidly‟ but also accurately. This is where approaches like Communicative

Language Teaching (CLT) are sometimes misunderstood. Wen Wu reports that

“one of the fundamental principles of CLT is that learners need to engage in

meaningful communication to attain communicative fluency in ESL settings”

(Wu 2008). Since fluency means also accuracy, it is clear that the aim of CLT is

to reach both.

Fluency is reflected mainly in two aspects: speed of delivery and

regularity, which means a natural amount and distribution of pauses (Bygate

2009). On the subject of appropriate placement of pauses Thornbury (2005)

says that:

Natural-sounding pauses | are those that occur at the intersection of clauses, | or after groups of words that form a meaningful unit. | (The vertical lines in the last sentence mark where natural pauses might occur if the sentece were being spoken.) Unnatural | pauses, on the | other hand, occur | midway between related groups of | words. (p. 7)

From my experience it seems, however, that the distribution of

pauses is completely omitted from LT. To investigate this a bit further I have

studied several EFL coursebooks 10 and have not found any formal instruction on

the issue of pauses distribution. Even though it could be that the level of

proficiency that the books under scrutiny focus on is too low, I still cannot

10 Cunningham and Moor (2008), Dellar and Walkley (2005), Hutchinson (2007), Kay (2008), Redston and Cunningham (2009) and Soars (2007).


remember this issue being addressed in any coursebook I have taught from

intended for higher levels. Although it may seem that teaching linking, stress

and rhythm the issues which are usually included in ELT are to a certain

extent linked to pausing, I would like to argue that appropriate placement of

pauses correlates more with the concept of tone units. They are the elements

that tell us exactly where pauses can be made.

Unfortunately, I was not able to find any studies to confirm my

belief that formal instruction in this aspect of fluency is recommended.

However, from my personal language learning experience I remember having

benefited especially from learning about this phenomenon. It was only when I

learned how tone units work in English that I first started to notice how native

speakers put their utterances one after another in speech and where they make

pauses. Having instructed my students in this I have also observed

improvement towards more natural distribution of pauses and linking of words

with some of them. Therefore empirical research in this area would be


Next, one of the issues that have been discussed in the previous

chapters is the importance of work with spoken data and transcripts in ELT. For

attaining fluency, the use of authentic texts and spoken data is significant.

Guillot (1999) reports that there are “practical as well as academic reasons for

making the study of spoken data native speaker and learner data an

integral ingredient of a pedagogy of fluency” (1999: 61). One of the reasons

that she lists is that “it can facilitate the emergence of individual paradigms of

fluency, enable students to identify the features and strategies of greatest


relevance to them as learners and communicators, and, concurrently, help them

to exploit both their strengths and weaknesses more efficiently” (p. 61). She

further states that:

) (

fluency more critically, can be used as a platform for helping learners to negotiate the shift from communicative control and sophistication, and project the development of their fluency beyond the confines of formal settings to transcend their inescapable limits time and restricted exposure to resources. (p. 62)

it provides a teaching and learning framework for approaching

In other words, Guillot supports my previous argument that use of

spoken data facilitates attaining fluency and that fluency leads to autonomy. As

she further puts it: “to teach fluency, in this sense, fits in with what Grenfell

and Harris describe as returning „ownership‟ of the language to learners” (1999:


Quite importantly, in the light of research evidence 11 , Thornbury

(2005) and Thornbury and Slade (2007) suggest that reaching native-like

fluency is only possible thanks to prefabricated chunks or formulaic language

that speakers use. These units include fixed phrases and idiomatic chunks such

as on the other hand, at the end of the day, or It‟s a small world. Johnson

(1996) states that a great deal of formulaic language is acquired unconsciously

either from direct transfer from L1 or from exposure to authentic L2 input. He

refers to these language items as „acquired output‟, i.e. language that is

11 David Wood summarizes a number of studies that confirm the importance of formulaic language for fluency, e.g. Raupach 1984, Sajavaara 1987, Pawley and Syder 1983, etc. (Wood



acquired unconsciously and produced automatically in tasks which require „high

automaticity‟. Oral communication due to its real-time processing demands falls

under this category.

There is evidence which supports the notion of acquired output,

particularly the importance of extensive exposure to authentic language input

for formulaic language acquisition and use. For example, Cock et al. found that

advanced speakers use prefabricated language less than native speakers and

for different pragmatic purposes (Thornbury and Slade 2007). There are several

other studies presented by Wood (2002) which suggest that language learners

use less formulaic language than native speakers due to limited exposure to

authentic input. In addition, Wood discusses further implications which these

findings have for LT. He recommends that classroom activities “consist of

exposure to large amounts of input, with attention paid to the formulaic

sequences being used” and stresses the importance of linking particular

formulas to particular pragmatic ends (2002: 10). Thornbury and Slade (2007)

making references to Lewis 1993 and Ellis 1998 and 2005 confirm that

acquisition is best achieved through massive exposure and explicit instruction.

In my experience, some of the greatest resources of formulaic

language which learners love to work with are songs and films (or series). Song

lyrics usually contain a good deal of chunks, idioms, etc., which can be

exploited in the classroom. One of the advantages of songs is that learners can

listen to them repeatedly inside and outside the classroom. If a song is

particularly catchy, learners are quite likely to memorise a great deal of text by

themselves. Apart from this, songs are good sources of authentic language,


allow learners to experience a variety of accents, motivate learning while

making it fun, break down barriers and build a relaxed atmosphere. Besides,

songs can be used with all ages and practically all proficiency levels if they are

chosen appropriately.

In terms of practicality, with the arrival of the digital era, songs can

be easily obtained for classroom use along with transcribed lyrics on the

Internet. ESL websites describe a number of activities which teachers can

exploit when using songs. My favourite is to write down language from song

lyrics (e.g. formulaic chunks) on little cards and stick them on the board. In the

classroom, learners form two rows. While listening to a song, they compete to

be the first to grab the card with the lyrics written on it, which they have just

heard. A variation of this awareness-raising activity is to prepare more sets of

cards and distribute them in pairs. While listening to a song, learners put the

words in the same sequence in which they hear them.

As far as visual input is concerned, there is a great range of activities

that can be done with films, series, soap operas, etc. For instance, a teacher

plays a muted scene of a film and puts a few lines or phrases from the scene

on the board. In pairs, learners try to make a dialogue which might be taking

place in the scene while inserting the language on the board into their

dialogues. After that, they watch the scene again but this time with the sound

on and check what was really said. Next, they can reconstruct the dialogue

based on what they heard and saw. Another variation of this is distributing sets

of cards with phrases from the scene on them in pairs, or groups depending on

the number of actors in the scene. Learners do a similar activity but this time


they grab a card with particular language whenever they use it in their

dialogue. The one who has the most cards wins.

Another good idea on how to practise formulaic language is to cut

out newspaper headlines which contain prefabricated chunks and ask learners

to speculate in small groups what the story behind the headlines might be and

agree on the version of a story which is most probable/interesting/unusual, etc.

This way learners use the language in the headlines meaningfully several times

while speculating about the stories. They also make use of a great range of

linguistic means when negotiating their stories. Next, learners can be asked to

present their ideas to the class. At the following stage, learners can be given

real texts which the headlines refer to. Discussing the stories with their partners

or within their groups, they compare how the texts differ from their own


Even though it has been said that fluency and accuracy are closely

linked together, for the purposes of LT, activities to practise speaking are

sometimes identified as fluency or accuracy focused. This is not to exclude one

of the two concepts from teaching but rather to point out what the main

purpose of the activity is, i.e. to concentrate mainly on using language

accurately or the ability to „get the message across‟. The former is usually used

to practise a particular linguistic phenomenon or language. The latter concept

usually entails using a broader range of skills and helps learners train strategic



Whether an activity is accuracy or fluency focused depends on a

particular task learners are asked to complete. For example, role plays can be

accuracy focused if learners are asked to use particular language or phrases

which have been introduced earlier in a lesson. However, the same task type

can be also fluency focused if learners are to act out roles that require a

broader range of knowledge and skills and the task is freer.

Typically, the more controlled a task is, i.e. requires learners to use a

fixed set of phrases or a certain formula, the easier it is for learners to

concentrate on a form and practise accuracy. Therefore tasks which require

only limited language are used as accuracy-focused activities. These include

certain types of information gap activities such as „Find someone who‟ (Activity

7), questionnaires, picture description, etc.

The advantage of controlled accuracy-focused activities is that they

can help learners use language which they are not ready to use yet by

themselves and thus gradually transfer passive knowledge into their active use.

The downside of accuracy-focused activities is that they practise only small

samples of language. Furthermore, if too structured or offering a certain

pattern to be used, they do not require learners to think much about the

meaning of what they say and can become meaningless. This point has already

been discussed in the previous chapter with the section on transactional and

interactional tasks (see section 2.1.6). Another fact is that if teachers use

accuracy-focused tasks for language which learners are not ready to use, the

activity will not bring much of a result.


When conducting accuracy-focused tasks it is important that learners

understand (and are instructed upon if necessary) that a particular activity is

accuracy focused. Otherwise a task can easily miss its aim. If learners do not

strive for accuracy but rather treat accuracy-focused tasks as a fluency practice,

their language skills will hardly develop further. This is mainly because

accuracy-focused tasks are designed to practise language in a very limited way.

Therefore it is crucial for teachers to make sure that learners understand the

real objective of an activity, albeit not stated explicitly at all times. In addition,

because of their limited language practice, accuracy-focused activities should

not be employed ad inifinitum at the expense of fluency practice.

All in all, for all accuracy-focused activities it is true that they should

be understood as transition tasks towards independent use of certain aspects of

language. It is therefore essential that teachers do not stop language practice

at the controlled level. The ultimate aim is to give learners opportunities to take

part in meaning-focused (rather than form-focused) activities. This can be done

at practically all levels through a gradual introduction of free tasks where

learners can utilise language that has been practised in a controlled or semi-

controlled way.

Unlike accuracy-focused activities, tasks focusing on fluency aim to

reflect natural language use, promote communication, require meaningful use

of language and produce language that may not be predictable. Thanks to their

lower level of predictability, they better reflect the requirements which are

placed on learners when producing spoken texts outside the classroom. On the

other hand, fluency-focused activities can be a real challenge for learners


especially if their overall proficiency is too low. The key to success in this case is

to introduce free tasks into tuition gradually, to choose tasks which learners are

likely to accomplish, to pre-teach necessary language and strategies and

possibly to demonstrate how a task can be successfully completed by giving

learners a brief example.

Typical fluency-focused task types are opinion-sharing activities like

discussions and debates, storytelling, creative tasks such as designing plans for

a new school facility and board games which require learners to speak on a

particular topic, etc. Apart from this, many accuracy-focused tasks may be

adapted to focus on fluency as well. For example, Activity 7, the „Find someone

who‟ task, can be made more complex by asking learners to expand their

answers if positive and report in detail about their experience, such as „the best

film they have seen‟ or „the most beautiful place they have visited‟. Naturally,

such a procedure is more time-consuming.

It should be added at this point that scholars differ in their

understanding of what is necessary for a communicative task to be effective.

For instance, first proponents of CLT suggested that three basic conditions must

be fulfilled in order for classroom communication to ensure progress in L2.

These are communicative purpose, information gap and language choice

(Littlewood 1983). By contrast, scholars supporting task-based approach

advocate L2 practice which is interactive, meaningful and includes a focus on

task-essential forms (Ortega 2007). To present contemporary trends of CLT, let

me use Brown‟s list of seven principles for designing speaking activities:



use techniques covering the spectrum of learner needs, i.e.

include both accuracy and fluency focused activities

2. provide intrinsically motivating techniques which appeal to

learners goals and interests

3. encourage the use of authentic language in meaningful


4. provide appropriate feedback and correction

5. capitalise the natural link between speaking and listening. Here

the author highlights the fact that language production is often

initiated through language comprehension. Therefore both

skills should be integrated in LT (see also Nation and Newton


6. give learners opportunities to initiate oral communication

7. encourage the development of speaking strategies

(based on Brown 2008)

To conclude, this section showed that it is necessary to utilise both

accuracy and fluency focused tasks in the classroom. Accuracy-focused

activities are likely to help learners use language correctly, while fluency-

focused activities will help them produce fluent stretches of language. On the

whole, fluency-focused activities are the ultimate goal in a classroom setting

because they help learners prepare for „what is out there‟. As has been pointed

out, another important aspect of teaching fluency is assisting learners in

building their formulaic language. Furthermore, this section mentioned some