Keywords

1 Introduction

In today’s English classrooms, we often see language learners sitting in pairs and groups talking or working together on a task. Such kinds of activities are aimed at helping learners gain confidence and fluency in speech and are particularly necessary in countries where English is not widely used. This practice is based on the assumption that through frequent practice with their peers, learners will transfer speaking skills from the classroom to real-life communication. This approach was identified in a review of speaking instruction by Burns (1998) as the indirect/transfer approach. Another approach she identified is the direct/controlled approach in which learners focus on getting the forms of the language right through direct instruction of grammar and pronunciation through drills, structure manipulation and consciousness-raising activities. This direct approach was common before Communicative Language Teaching (CLT) methods became influential in many parts of the world resulting in the currency of the indirect/transfer orientation in speaking activities.

On the whole, there have been no perceptible paradigm shifts in methods and practices for teaching speaking (Goh and Burns 2012; McCarthy and O’Keeffe 2004). Nevertheless, new understandings of the relevance of discourse analysis and features of spoken English have resulted in an expansion of the scope of the direct/controlled approach. This expanded approach, informed in part by genre theory, introduces learners to a variety of spoken texts and their respective discourse structures through direct teaching of language and discourse knowledge needed for successful oral communication (Burns 1998). Additionally, corpus research work such as the CANCODE spoken corpus (McCarthy and Carter 1995) has provided evidence of significant differences between spoken and written English and their pedagogical implications (Carter 1998). There are compelling reasons why a methodology for teaching speaking today would need to move away from a model based entirely on written language to ensure that the language that second language (L2) learners develop is natural and reflects the way English is spoken in real-life (McCarthy and Carter 2001), a view that is also shared by some practitioners (Goh 2009; Timmis 2005).

Taking into consideration the pedagogical landscape for speaking instruction, this chapter proposes a comprehensive and holistic approach, which integrates the combined strengths of direct and indirect instruction with the power of learners’ metacognition (cognition about thinking and learning processes). This approach is further informed by some recent research findings on pedagogical processes that can scaffold the development of L2 speaking. Within this discussion, I will explain the construct of speaking and highlight pedagogical procedures that can contribute positively to speaking performance. I will present the implications of such understandings and suggest pedagogical principles that can enhance current practices for facilitating second language speaking development in and beyond the language classroom.

2 Second Language Speaking

Understanding what speaking entails is essential if we are to teach it well. We often say someone is a good speaker because that person speaks confidently, fluently and grammatically. In some learning contexts, a person may be considered a good speaker if he or she sounds like a speaker from one of the traditional native-speaker countries such as the UK or the USA. Some people may say that a good speaker is someone who is able to influence others with his or her words. While ‘good’ speaking may seem such a self-evident phenomenon, the construct of speaking is anything but simple. In this section, we will examine what L2 speaking is by discussing the concept of speaking competence and the processes involved in speech production. This is followed by selected research highlights that offer pedagogical procedures that can potentially enhance L2 learners’ speaking performance.

2.1 Speaking Competence

Speaking involves dynamic interactions of mental, articulatory and social processes. To express a message, speakers need to decide what to say and use their linguistic knowledge to construct utterances and encode this message in sounds and sound patterns that can be recognised and understood by their listeners. They also need to consider the context of interaction and engage their listeners in socially appropriate ways through various linguistic choices and forms. For example, speakers may choose to use certain vocabulary or register when speaking with people with whom they have shared knowledge and experience. Speaking is also influenced by varied cognitive and affective factors, such as the ability to process speech quickly and feelings of anxiety respectively. To begin our discussion, it is instructive to examine a description of L2 oral communication by Johnson (1981, p.11) that is still very much relevant today:

Consider for example what is involved in producing a conversation utterance. Apart from being grammatical, the utterance must also be appropriate on very many levels at the same time; it must conform to the speaker’s aim; the role relationships between the interactants; to the setting, topic, linguistic context, etc. The speaker must also produce his utterance within severe constraints; he does not know in advance what will be said to him (and hence what his utterance will be a response to); yet, if the conversation is not to flag, he must respond quickly. The rapid formulation of utterances which are simultaneously “right” on several levels is central to the (spoken) communicative skill.

Johnson identifies critical aspects of L2 speaking that are elaborated in the sections that follow.

2.1.1 Enabling Skills

An important characteristic of competence is the ability to produce utterances that are grammatically accurate, a notion we will return to later. Accuracy alone, however, is insufficient. Competent speakers need to use language for myriads functions so as to achieve a range of communication goals. They do this through various sub-skills that enable them to navigate the social elements at work in any interaction so that what is said is not only clear but also appropriate to the context and acceptable to their listeners. To do this, they need to determine what type of information and how much of it is needed, as well as effective ways to express their meaning, organise their speech and articulate the sounds that accompany their speech intelligibly. The centrality of skills in the conceptualisation of speaking competence is demonstrated in various discussions of the construct of speaking in which a number of production and interaction skills have been identified. Goh and Burns (2012) have grouped speaking skills into four sets or clusters of skills, each with many sub-skills respectively that are appropriate for the learning and communication needs of learners:

2.1.1.1 Pronunciation Skills

These skills that are articulatory and phonological in nature enable speakers to produce sounds at the segmental and suprasegmental levels. At the segmental level, learners need to articulate discrete sounds such as vowels, consonants and diphthongs, and clusters of these sounds through movement with and inside of their mouths to produce intelligible sounds through the articulatory tract. The suprasegmental level concerns overall sound patterns of utterances or parts of an utterance and are realised mainly but not exclusively through prominence (stress of selected syllables in key words) and tones (pitch movements in selected key words). Suprasegmental features are not mere reproduction of sentence stress patterns to show attitude or emotions, as suggested by some instructional materials for pronunciation. Instead, they have important communicative value and are produced in response to the real-time unfolding of meanings in discourse during any interaction (Brazil 1985/1997). Consider the following example:

A: Where’s my bag?

B: //UNder the TAble//

A: (Looking at the top of table) Where?

B: //NOT ON the table// UNder it//

B’s first reply shows what would normally be considered as “correct” stress pattern because the key content word ‘table’ is given the more prominent stress. In B’s second reply, however, the stress is not on the word ‘table’ because this information is no longer new but is given or shared. Instead, the stress is found in grammar words in the form of the prepositions ‘on’ and ‘under’ where greater prominence has been assigned to show the contrast in the location of the bag.

2.1.1.2 Speech Function Skills

We use speech to perform speech acts, that is to say we produce spoken language to get things done. To achieve this, speakers need to produce utterances that can convey desired communicative functions through a combination of appropriate language use, vocabulary choice and grammar. Inventories of language functions for speech can be found in many skill-based language syllabuses or documents such as the Common European Framework Reference that specify the competencies that language learners are expected to achieve at various levels of proficiency. There are many basic language functions that learners need to show, for example, inform, accept, decline, request, explain and describe. Individual learners’ functional repertoire will depend largely on their contexts of interaction and the purposes for which speech can fulfil. Compared to young learners, adult learners in academic or professional situations would need to convey more complex functions such as negotiate, advise and argue.

2.1.1.3 Interaction Management Skills

Some speech functions are directly related to the ability to manage an interaction or regulate the flow of conversations. Just as children learning their first language need to learn how to initiate and sustain face-to-face interactions, language learners need to develop skills to do so in another language. These include but are not limited to initiating an interaction or conversation, taking turns, giving turns, asking for clarification, changing topics and closing an interaction. Adult learners’ prior experience would allow them to understand the moves needed in face-to-face interactions, but they still need to learn to use the language to convey these moves. Formulaic expressions for indicating the specific functions are an important part of learners’ repertoire of interaction management skills. Moreover, because of cultural differences, language learners will also need to recognise their interlocutors’ moves as well as creating moves and utterances in socioculturally appropriate ways themselves.

2.1.1.4 Discourse Organisation Skills

Most spoken interactions occur in contexts where participants have equal or similar opportunities to talk. Very often, however, language learners may have longer turns and are required to produce extended pieces of discourse, for example, when giving a presentation, explaining or describing procedures and narrating an event or a story. They will therefore need skills to construct these spoken texts in ways that are consistent with the sociocultural conventions for the respective genres in the language being learnt. In addition to knowing about discourse routines (the stages and moves that are typically found in specific contexts), learners need relevant language to frame the moves. For example, in giving a presentation, learners must make use of discourse markers to signpost transitions. These markers can be simple such as using the word ‘Next’ or complex such as including a summary of what has been said and progressing to the next ‘We have just examined X, let’s now consider Y.’ Young learners learning to tell a story in the target language will need to know the structure of a narrative (orientation-problem-resolution-coda) and use markers to indicate these transitions.

2.1.2 Communication Strategies

Communication strategies are special techniques that learners need to employ during oral communication. They can have a social function for enhancing interaction or a psycholinguistic function that compensates inadequate vocabulary and other language-related problems (Nakatani and Goh 2007). Given the constraints of time and inadequate language mastery, learners also often need to employ communication strategies to keep the conversation going or to prevent flagging (Dörnyei 1995). For example, learners may use interactional strategies such as asking for clarification or repetition and comprehension checks before responding to their interlocutors to ensure that they can give a correct response and gain time while formulating a response. Less proficient learners who do not understand what they hear and are unable to express their meanings immediately may ask for assistance directly. They may also adjust their message according to their competence by reducing what they say to the minimum or steering the conversation away to a new topic which they are more familiar with.

Learners may also use formulaic expressions or discourse markers, such as ‘Well’, ‘Yes, that’s a good point’ as hesitation devices to gain more thinking time, and use generic terms or vague words such as ‘thing’ to substitute a more precise term which they do not know in the target language. This last strategy is also called approximation and is an example of cognitive strategies used for solving problems when L2 speakers encounter gaps in lexical knowledge and related linguistic problems. Other cognitive strategies include paraphrase, circumlocution, word coinage and borrowings from L1. Learners can also use metacognitive strategies to plan what they want to say, self-monitor during speaking and evaluate their language and message after speaking (Bygate 1998). Given the position of English as an international language, learners must also develop strategies that enable them to communicate across cultures (see Newton, this volume).

2.1.3 Language and Discourse Knowledge

The notion of grammaticality in Johnson’s (1981) observation was concerned mainly with syntactic (word order) and morphological features such as verb inflections, and noun plurality. In recent decades, however, our understanding of grammar has expanded to include knowledge and use of grammar in relation to spoken genres as well as structuring different kinds of spoken genres, i.e., types of texts produced in different communicative events, such as conversations, lectures and interviews, according to the sociocultural context (Burns 1998). Speakers need to use grammar that supports the production and organisation of the respective genres. For example, in producing oral narratives or stories, various forms of the past tense are most common whereas giving instructions or directions (e.g., procedural texts) will require the use of the imperative forms of verbs. Another extended notion of grammaticality is the speakers’ knowledge of and ability to use spoken grammar, as it would no longer be possible to ignore the compelling evidence from spoken language data in any discussions of speaking pedagogy (McCarthy and O’Keeffe 2004).

3 Processes in Speech Production

Our understanding of L2 speaking has been informed by useful models of speech production in cognitive psychology. One model that has been adopted in several L2 speaking discussions is Levelt’s (1989) framework of conceptualisation, formulation and articulation based on first language speakers (see Bygate 1998). Conceptualisation is a speaker’s mental planning process to determine what he or she intends to say. Information is selected and intentions of speech acts are activated at this stage. Such a mental concept or plan may exist as a general idea, but the message still has to be expressed in relevant words that are strung together grammatically. This requires the accessing and retrieval of vocabulary that is stored in long term memory existing as individual words, phrases or even complete chunks of utterances as well as the application of grammar knowledge of the language. During the process of formulation, speakers will actively draw on their knowledge of the language to express their meaning as clearly and precisely as possible. The actual expression of the ideas for the listeners occurs when the words are said aloud through phonological encoding at the segmental and suprasegmental levels. This physical process which is called articulation is brought about by the activation and control of components of the articulatory system.

3.1 Directions of Speech Processing

According to Levelt (1989), while the processes of conceptualisation, formulation and articulation often occur interactively, they can also take place in a linear manner. This is to say that one process may occur while another is still taking place, but it is also possible that speakers may engage with the processes separately before speech acts are performed through a demonstration of the individual or collective functions of the utterances. Interactive speech processing occurs more commonly in spontaneous speech production where speakers have to decide what to say, how to say it and say it aloud. In L2 speaking this also presents the greatest challenge for learners and they may have to resort to communication strategies to buy time as we have discussed previously. They may also have to process their speech in a linear manner such that one process (for example, articulation) occurs only after another (for example, formulation) has completed.

3.2 Metacognitive Processes

In addition to these cognitive and articulatory processes, speech production also frequently involves metacognitive processes. These are mental processes operating at a level beyond the direct manipulation of language and ideas. Metacognitive processes manage and regulate speech as it is processed cognitively and articulated physically. A primary metacognitive process is monitoring (Bygate 1998). This occurs when speakers check the accuracy and appropriateness of what is being said and how it is being said all the time when they are saying it. Another metacognitive process is evaluation which takes place following speech production. Speakers may review what they have just said and decide whether they have been effective in conveying their thoughts, ideas or information and the achievement of their communication goal. This may occur immediately after an utterance is articulated or at the end of a speech event. Another key metacognitive process is planning and this may overlap with the conceptualisation phase in situations when speakers have plenty of time to think about what they want to say, for example in preparing for a presentation.

3.3 L2 Speaking Performance

Although language learners also engage in similar processes of speech production, they encounter various challenges that can affect their speech fluency. To explain L2 speaking performance, a multidisciplinary, cognitive science framework was proposed by Segalowitz (2010) that is informed by neurocognitive science and social psychology of bilingualism. It explains L2 speech performance in terms of the dynamic relationships among a number of variables or sources, which can variously exert demands on L2 learners’ speech. These are cognitive perceptual systems that underlie speech production, utterance fluency features (e.g., speech rate, hesitation and pausing), motivation (e.g., willingness to communicate, beliefs, language and identity, and the concept of L2 self), the social or interactive communicative context, and fluency-relevant perceptual and cognitive experiences (e.g., exposure, opportunities for repetition practice). L2 fluency is therefore affected by many demands, such as a limited cognitive processing capacity because conceptualisation, formulation and articulation need to take place within constraints of limited content, language and discourse knowledge. Some learners may also be hampered by inadequate cultural knowledge that can otherwise enhance their oral communication and enhance their confidence when talking with English speakers from other countries.

In a face-to-face communication situation where there are time pressures to ‘perform,’ learners will focus more on conveying the meaning of what they want to say rather than worry too much about the accuracy of their language (Skehan 1998). In other words, many learners may not have the luxury of time and processing capacity to monitor what they say constantly. This said, language learners do try to check on what they say whenever possible in order to enhance their performance, and their self-monitoring processes are evident in the presence of self-repairs. Learners do notice their mistakes or lack of clarity and correct themselves. At the same time, language learners also employ communication strategies to seek assistance, gain extra time or improve what they say. The ability of language learners to maintain interaction no matter how challenging this proves to be as well as compensating for a lack of lexical knowledge shows they also engage in the metacognitive processes of planning, self-monitoring and evaluation.

4 Speaking Tasks

Teachers plan a range of speaking tasks with various levels of demands and outcomes to give learners opportunities to practise their spoken English. Some of these tasks may require learners to talk together in groups to arrive at a solution to a given problem while others may simply require them to exchange specific information. There are broadly three types of speaking tasks that encourage genuine communication among learners: communication-gap tasks, discussion tasks, and monologic tasks (see Goh and Burns 2012 for details). In communication-gap and discussion tasks, learners interact with a partner or others in small groups to convey information and viewpoints to achieve a communicative outcome. There are many forms of ‘gaps’ in communication-gap tasks and these include missing information or details which one learner will have to describe, narrate or explain to their partner.

In comparison, discussion tasks create an even more authentic context for speaking and interaction because learners share their personal views with one another. When they have to discuss an open or controversial topic, for example, learners can draw on their own background knowledge, experience and beliefs. When a consensus or solution is required, they will have to negotiate with one another for an outcome that everyone can agree on. Sometimes, group discussions can also occur through simulations, which are classroom activities that reproduce or create a situation that is close to real life concerns. In simulations, learners are given scenarios in which they take on a role, such as a doctor, a Member of Parliament, a school counsellor, and a parent to discuss an issue with others taking on other roles.

In contrast to the two kinds of tasks just mentioned, monologic tasks require learners to present ideas, information and views individually to a single listener or a group of listeners. For example, they may give a talk, tell a story or present a report. They may also speak extensively on a topic or a theme without interruptions. They may be asked to give spontaneous and unedited talks or planned and rehearsed ones. These ‘performances’ can be done in front of the whole class, but doing them in small groups is preferable because it reduces anxiety for the speakers and enables peers to ask questions and give feedback in a less threatening environment. Teachers can plan different kinds of monologic tasks and vary the duration of the monologue according to learning objectives.

5 Enhancing Second Language Speaking Performance

Speaking in a second language clearly presents many challenges to language learners. These challenges, however, do not always get addressed in the classroom. Although students have opportunities to develop their confidence and fluency through oral activities, they do not in general receive much of the scaffolding they need for learning and improvement to take place during the instructional process. Spurred by their motivation to succeed, many learners may put in extra time and effort to practise their spoken English by using self-study techniques or seeking opportunities to speak with more competent speakers of English. Some learners may find their progress slow while others may feel that they do not learn enough by just practising in class with peers who are not any better than themselves. These are genuine concerns, and there are ways for teachers to support learners and help them succeed. Recent research has provided new understandings about cognitive and general learning processes for L2 learners, and these understandings can provide further directions in the way we enhance speaking pedagogy. I discuss below three strategies that can enhance L2 learners’ speaking performance.

5.1 Pre-task Planning

Some researchers have investigated whether it was useful to give learners time to plan and prepare for a speaking task and how pre-task planning might have an impact on their fluency, accuracy and language complexity (see for example, Skehan and Foster 1997, 2005). Varying degrees of positive effects have been reported for all three dimensions of speech but the effect on accuracy is still inconclusive. Another type of pre-task planning focuses on the strategies that learners could use during the task to facilitate communication and intelligibility. Strategy training conducted at the pre-task stage enabled some learners to apply strategies during speaking and produce speech that was significantly more fluent than that of learners who did not receive any training. In addition, pre-task planning time has allowed some test-takers to prepare themselves for a speaking test by using language-related strategies as well as strategies for content and discourse organisation (Wigglesworth and Elder 2010). Some researchers speculated that while pre-task planning was helpful for learners, individual differences such as the ability for self-monitoring and repairs could confound the effects of pre-task planning, so other ways of helping learners improve their speech production such as task repetition should be explored.

5.2 Task Repetition

Task repetition is the repeated use of the same or similar communication task or discourse sequences by learners with the same or different people (Bygate 2001). Research has shown that when learners repeated a speaking task they produced more accurate and natural speech and demonstrated better framing of their narratives (Bygate and Samuda 2005). When repeating presentations to different audiences, learners integrated lexical knowledge generated from the first task, showed a wider range of lexical items and increased their accuracy in grammar and pronunciation (Lynch and Maclean 2000, 2001). Allowing learners to repeat a task can free up valuable cognitive space for learners which would otherwise be severely taxed by the need to attend to different aspects of their performance. For example, when task repetition was combined with a form-focused activity, learners were able to direct their attention more effectively at form in the repeat performance (Hawkes 2012).

5.3 Metacognition Enhancement

Metacognition is an individual’s ability to think about his or her own thinking and learning. It encompasses knowledge of one’s own learning (person knowledge), the nature and demands of learning tasks (task knowledge) and how to approach these tasks (strategy knowledge), and the actual use of strategies for problem-solving as well as monitoring, regulating and orchestrating thinking and learning processes (Flavell 1976). The role of metacognition in learning has been discussed extensively in educational psychology. In L2 speaking, it has been examined specifically in two areas. The first is the use of communication strategies as previously explained and the other is the development of learners’ metacognition about speaking through awareness-raising and strategy-instruction activities (Goh and Burns 2012). A recent study reported that a group of learners’ metacognitive knowledge about speaking improved substantially when they were given the Common European Framework of Reference (CEFR) level descriptors for speaking to support their learning (Glover 2011). The learners were also able to use the descriptors effectively for self-evaluation of their speaking development. In another study involving learners of Chinese as a second language, the learners’ speaking improvement was attributed partly to the use of metacognitive reflections in an intervention programme (Tan and Tan 2010). One of the things that the learners did in this programme was evaluating, monitoring and planning their speaking performances. Improvements in the pronunciation of a group of EFL learners were also attributed to their engagement in metacognitive processes such as weekly journaling (He 2011).

6 A Comprehensive and Holistic Approach

There are many good teaching practices for speaking today. Although valuable and useful, they do not adequately offer scaffolding processes that allow language learners to benefit more extensively from time spent in and out of class. This limitation therefore calls for an enhanced approach that is guided by a coherent understanding of the construct of L2 speaking, how relevant research findings can inform pedagogy and the potential of metacognition for language learning tasks in and outside the classroom for speaking development. Such an approach addresses these ideas comprehensively and responds to learner needs holistically. Based on earlier discussions about L2 speaking, a number of implications and pedagogical principles can be drawn for such an approach. These are presented in Table 1 below. These considerations are relevant for the traditional face-to-face speaking classes as well as for speaking practice activities that are technology-enabled.

Table 1 Implications and principles for teaching speaking
Full size table

Controlling the accuracy of language that learners produce by pre-teaching them the forms and structure of language and discourse is not always effective. Neither is it adequate to just plan practice activities in the hope that our students will eventually transfer skills and knowledge from classroom activities to fluent communication beyond the classroom. The points presented in Table 1 illustrate ways in which teaching speaking needs to take a comprehensive and holistic approach. This approach combines the strengths of direct/controlled and indirect/transfer ways of teaching speaking, and integrates them with supportive metacognitive processes to provide learners with maximum benefits for speaking development. By doing this, we make the processes of speaking and learning to speak more visible to our learners. The objective of speaking instruction is to help learners develop the fluency and accuracy of expert speakers who can convey their message clearly and effectively in socioculturally appropriate ways. Just as discourse analysis and conversation analysis have prompted a renewal of the direct approach (Burns 1998), understandings about how cognitive and learning processes in L2 speaking can be supported can influence the way speaking instruction is carried out for the future.

An issue with speaking instruction is the transience of spoken language. Teachers seldom have a record of what their students say, especially when they are talking in groups and the teachers have to walk around the class to monitor what is said and help students with vocabulary and other things. More importantly, students themselves do not have a record of what they have said. This lack of permanence in learners’ speech production hampers opportunities for noticing and analysis, two important processes in learning. It is important therefore to find ways of giving transient spoken language some permanence through the affordances of technology. For example, students can record their speech on their smart phones for review at a later time. The audio recordings can also be uploaded to a common platform used in the school or institution. Equally important is for teachers to make the speaking process visible to learners by giving the learners opportunities to focus on the knowledge and language that support the skills needed to accomplish a task as well as the strategies that may be needed to overcome limitations in their abilities. As video recordings are widely available nowadays on the internet, teachers can look for suitable recordings of expert speakers doing a similar task, such as giving a talk, or participating in an interview or a discussion. These can be used to show learners the way specific language and discourse items can be used to enhance effectiveness. Teachers should also embed within speaking lessons procedures such as pre-task-planning and task repetition which research shows can be beneficial to learners. These pedagogical processes, however, are still rare in many speaking classrooms.

Most teachers would agree that learners can benefit from getting extra time to prepare what they have to say. Preparation is believed to help them be more fluent in their speech, use more appropriate vocabulary and generally become more grammatical in their production. If nothing else, the content of the speech or utterances will be expected to be richer because the students will have time to gather their ideas about what they want to say. This would enable learners to monitor their speech and do self-repairs when necessary to enhance the clarity of their message and the accuracy of their language. Including task repetition as a pedagogical procedure gives learners a second chance to improve their performance after the ‘rehearsal’ when the task was first carried out. When learners do a speaking task just once, they typically do not give a complete and polished performance; mistakes are common. By repeating a task, they get a chance to integrate knowledge constructed in the first attempt into the repeat performance. They also get the benefit of evaluating their own performance and becoming more aware of the nature and demands of the task based on their prior experience. In most classrooms, learners do a speaking task before they move on to other language learning tasks for reading and writing. In some situations, speaking is seen as a pre-reading or pre-writing task instead of a learning task in its own right. By asking learners to repeat a task, teachers are highlighting to students that they are not just doing a speaking task but learning how to speak.

The approach presented in this article can be seen in the Teaching Speaking Cycle (TSC) by Goh and Burns (2012). In the TSC, learners develop their speaking through a number of activities and tasks as the teacher guides them systematically through each stage of the cycle. It engages learners, individually and with peers, through planned reflective processes, oral practice with selected types of speaking tasks, activities for noticing and analysing language and discourse, repetition of the speaking task as well as input and feedback from teachers and peers. The seven stages in the TSC consist specifically of the following: focusing learners’ attention on speaking, providing input and/or guide planning, conducting speaking tasks, focusing on language/discourse/skills/strategies, repeating speaking tasks, directing learners’ reflection on learning and facilitating feedback on learning.

7 Conclusion

To teach speaking is to facilitate our students’ understanding of speaking processes and scaffold their development of speaking competence in a systematic and theoretically-principled manner. Speaking instruction should be more than putting learners in pairs and groups and giving them opportunities to communicate with one another in the target language. Giving students opportunities for practice does not automatically translate into learning the skills and language necessary for speaking effectively. Setting up an activity for oral practice is only one aspect of teaching learners how to speak. While the transfer or indirect approach has communicative authenticity, it needs to be enhanced so that our students can benefit directly from scaffolding processes for maximum learning to occur in each lesson. At the same time, we need to exploit the strengths of the direct approach to enable learners to understand the language and discourse as well as skills and strategies that they need in order to gradually become effective L2 speakers. Last but not least, we need to recognise that learning to speak in another language can create a great deal of anxiety for our students. They may also feel discouraged when they do not see improvements or are confused as to how they can manage their own learning processes.

By acknowledging that students need to engage with their learning beyond the cognitive and affective dimensions of speaking, teachers can provide them with the kinds of support that are lacking in many speaking classrooms. An enhanced conception of speaking pedagogy involves planning of activities which are underpinned by metacognitive processes and which not only allow learners to practise using oral language but also focus their attention on important linguistic elements and oral communication processes that can further improve their performance. Such an approach can offer rich opportunities for practice while making the development of speaking competence a visible process that learners can increasingly regulate and control. It offers a speaking pedagogy that engages our students’ thinking, action and emotions.