Oral Fluency: The Neglected Component in the Communicative Language Classroom

Abstract

In this paper, we argue that current instructional ESL resources must be supplemented to facilitate the effective development of learners' oral fluency. We summarize some of the pertinent literature on L2 fluency and report the results of a survey of fluency activities (free production, rehearsal/repetition, consciousness-raising, and use of formulaic sequences and fillers) found in 28 ESL learner texts and 14 teacher resource materials. The findings indicated a heavy emphasis on free-production tasks in both learner and teacher resource books, with less focus on the use of formulaic sequences, rehearsal, and repetition. Learner texts were sorely lacking in consciousness-raising activities; furthermore, fewer than half of the teacher resource books included these. We describe types of oral fluency instruction that can be integrated into L2 classes to address these deficiencies. Finally, we propose contact activities to assist learners in developing fluency outside their ESL courses, and suggestions for research.

Résumé

Les auteurs font valoir que les ressources actuelles en matière d'enseignement de l'anglais langue seconde (ELS) doivent être complétées afin de faciliter le perfectionnement efficace de la compétence orale des apprenants. Le présent article résume une partie de la documentation pertinente sur la compétence en langue seconde (L2) et présente les résultats d'un sondage sur les activités en matière de compétence (production libre, répétitions, sensibilisation, et utilisation de séquences comportant des formules et des espaces à remplir) trouvées dans 28 textes d'apprentissage en ELS et 14 documents sources d'enseignement. Les résultats ont montré un fort accent sur les tâches de production libre, dans les documents sources tant de l'apprenant que de l'enseignant, avec moins d'insistance sur l'emploi de séquences comportant des formules et sur les répétitions. Les textes des apprenants affichaient un manque important d'activités de sensibilisation, et moins de la moitié des documents sources des enseignants en comportaient. Les auteurs décrivent les types d'éducation en matière de compétence orale pouvant être intégrés au cours de L2 afin de remédier à ces déficiences et, enfin, ils proposent des activités de contact pour aider les apprenants à perfectionner leurs compétences à l'extérieur de leurs cours ELS et des suggestions pour la poursuite de la recherche.

In immigrant-receiving countries such as Canada, oral fluency is of great importance to second language (L2) learners who need to be able to participate in academic, occupational, and social contexts. Many learners invest enormous amounts of time and, in some cases, considerable financial resources in an effort to acquire English language proficiency, with the expectation that they will eventually be able to speak the language fluently. It is discouraging for learners who have made these commitments to struggle with oral fluency even after one or two full years of English as a second language (ESL) instruction (Derwing, Munro, & Thomson, 2008).

In this paper, we will review some of the existing research on L2 oral fluency development and examine published resources commonly available to adult ESL programs to determine to what extent they include fluency development tasks and activities. Then we will suggest some specific research-supported approaches to structuring tasks and instructional procedures. Finally, we will argue for a balanced approach to developing oral fluency in the adult ESL classroom.

Oral fluency is one of the most salient markers of proficiency in a second language. The term 'fluency' has a range of meanings, the most common of which is related to 'high proficiency,' that is, an excellent grasp of the vocabulary and grammar of a language. In this paper, we define fluency as a performance phenomenon related to 'flow, continuity, automaticity, or smoothness of speech' (Koponen & Riggenbach, 2000, p. 6).

Oral fluency has been shown to be correlated with appropriate speech rate (e.g., Derwing, Rossiter, Munro, & Thomson, 2004; Freed, 1995; Kormos & Dénes, 2004; Lennon, 1990; Riggenbach, 1991; Rossiter, 2009; Towell, Hawkins, & Bazergui, 1996); length, frequency, and distribution of silent pauses and non-lexical fillers, such as um and uh (e.g., Derwing, Rossiter, Munro, & Thomson, 2004; Foster & Skehan, 1999; Freed, 1995; Lennon, 1990; Riggenbach, 1991; Rossiter, 2009); and mean length of run (i.e., the average number of syllables between pauses) (e.g., Ejzenberg, 2000; Lennon, 1990; Riggenbach, 1991; Towell, Hawkins, & Bazergui, 1996). In the L2 classroom, oral fluency can be enhanced by the use of such linguistic features as formulaic sequences (i.e., high-frequency multi-word units such as by the way, in other words) (e.g., Boers, Eyckmans, Kappel, Stengers, & Demecheleer, 2006; Ejzenberg, 2000; Nattinger & DeCarrico, 1992; Towell, Hawkins, & Bazergui, 1996; Wood, 2001, 2006, 2009; Wray, 2002), and discourse markers (Guillot, 1999; Nattinger & DeCarrico, 1992; Tyler, 1992). Fluency can also be facilitated by pedagogical procedures such as consciousness-raising (to raise awareness of fluency features, [e.g., Boers, Eyckmans, Kappel, Stengers, & Demecheleer, 2006]); the provision of pre-task planning time (Foster & Skehan, 1996, 1999; Mehnert, 1998; Skehan & Foster, 2005; Tavakoli & Skehan, 2005; Yuan & Ellis, 2003); the imposition of time constraints on production (e.g., Arevart & Nation, 1991; Nation, 1989); and task repetition (e.g., Bygate, 2001; Gatbonton & Segalowitz, 2005; Lynch & Maclean, 2001; Nation, 1989).

However, as Gatbonton and Segalowitz (2005) suggest, although many communicative language teaching (CLT) classrooms promote general fluency, they do not provide the repetition necessary to achieve automatic fluency:

Although one component of fluency is automatic, smooth, and rapid language use, there are no provisions in current CLT methodologies to promote language use to a high level of mastery through repetitive practice. In fact, focused practice continues to be seen as inimical to the inherently open and unpredictable nature of communicative activities. Thus, when teachers believe that learning has reached the point where reinforcement of new forms through practice is necessary, they tend to revert to non-communicative means for attaining this end (such as pattern practice).

(p. 327)

In addition to a lack of meaningful and repetitive practice, many ESL classes offer little or no explicit, focused instruction on the development of oral fluency skills. For example, Derwing, Munro, and Thomson (2008) assessed the fluency of two groups of adult ESL learners over a period of two years. The researchers found that the Mandarin speakers in the study showed no significant improvement in fluency despite spending 25 hours a week in ESL classes for at least a year. The oral fluency of the Slavic language speakers, on the other hand, did improve significantly over time. The researchers suggested that the difference was related to the amount of exposure to English that the learners received outside of their ESL courses. They recommended that oral fluency be explicitly taught in L2 programs, particularly in view of the fact that some ethno-cultural groups may not have extensive exposure to English in the larger community. Other factors, such as motivation and ethnic affiliation, also affect fluency development, as Segalowitz, Gatbonton, and Trofimovich (2009) have pointed out, but nonetheless, instruction remains a key element in promoting ultimate fluency. Because many programs rely heavily on textbooks for ESL instruction, the present study was designed to determine the extent to which instructional resources provide support for the development of oral fluency in the ESL classroom, and to recommend fluency tasks that instructors can use to supplement existing materials.

Materials survey

We restricted our analysis to books that were available in Edmonton, Alberta, and that were used frequently. The learner textbooks (designed for low to high intermediate proficiency learners) and teacher resource books selected for analysis included: (a) 14 student textbooks that explicitly claimed to promote the development of oral fluency (according to publisher advertisement, synopses, tables of contents, and titles, e.g., Bridge to Fluency: Speaking; World Links: Developing English Fluency); (b) 14 general texts that were widely used in local Edmonton ESL programs (e.g., Touchstone; Canadian Snapshots); and (c) 14 teacher resource books that addressed the teaching of oral fluency (e.g., Fluency and its Teaching; Discussions that Work: Task-Centred Fluency Practice). Two textbooks from the Canadian Snapshots series were surveyed, but there were no other multiples of a given series.

We surveyed each book to determine if it included any of each of the following oral fluency activities:

1. a. consciousness-raising tasks (i.e., to raise awareness of fluency features) (Boers, Eyckmans, Kappel, Stengers, & Demecheleer, 2006);

2. b. rehearsal or repetition tasks (Bygate, 2002; Gatbonton&Segalowitz, 2005; Lynch & Maclean, 2001; Nation, 1989);

3. c. the use of formulaic sequences (Boers, Eyckmans, Kappel, Stengers, & Demecheleer, 2006; Ejzenberg, 2000; Nattinger & DeCarrico, 1992; Towell, Hawkins, & Bazergui, 1996; Wood, 2006, 2009; Wray, 2002);

4. d. the use of discourse markers (lexical fillers such as so; you know) (Guillot, 1999; Nattinger & DeCarrico, 1992); and

5. e. communicative free-production activities (e.g., general speaking tasks without a specific focus, traditionally seen as fluency builders in L2 classrooms).

The first four types of oral fluency activities we surveyed are frequently referred to in research and have been shown to enhance fluency; the fifth, free-production tasks, is traditionally the most common speaking activity in ESL classes.

Of the 14 student books that were purported to promote oral fluency, 86% included free-production activities, 71% gave explicit attention to formulaic sequences, and 50% included some form of rehearsal or repetition (see Table 1). These texts did not contain consciousness-raising activities, nor did they have a focus on fillers. Of the other 14 widely used student books, 86% included free-production activities, 64% gave explicit attention to formulaic sequences and some form of rehearsal or repetition, and 36% incorporated consciousness-raising tasks and a focus on fillers. We found that if learner textbooks had a specific type of fluency task, they tended to have several activities illustrating only that type of task. Only two of the student books surveyed, Speaking solutions (Matthews, 1994) and Touchstone I (McCarthy,

Table 1.

Student texts and teacher resource books that incorporate fluency activities

McCarten, & Sandiford, 2005) included all five types of fluency activities.

In contrast, the 14 teacher resource books provided a comprehensive repertoire of fluency development activities and teaching techniques. Free-production activities were again most common (in 71% of resource texts), followed by attention to formulaic sequences (in 64%) and rehearsal/repetition (in 57%). However, a greater percentage of instructor resource books (43%) contained consciousness-raising activities and use of fillers than did the learner materials. New ways in Teaching Speaking (Bailey & Savage, 1994) was the only teacher resource surveyed that represented the full spectrum of fluency tasks.

Overall, it appears from the survey that student texts rarely contain consciousness-raising activities and explicit instruction in the use of fillers; therefore, the instructor is responsible for supplementing learner materials with these types of activities. While a greater proportion of teacher resource books addresses these oral fluency approaches, these books may be less frequently utilized than student texts.

Priorities for developing fluency

The first step in developing instruction to promote fluency is to assess learners' oral productions to determine if fluency training is warranted and, if so, which aspects of fluency should form the focus of instruction. A checklist could be employed, using a scale to rate the effective use of pause length, frequency, and location, and speech rate. This should be done with different types of tasks (e.g.,monologues, dialogues; structured, unstructured tasks) to provide a complete representation of the learners' abilities (Ejzenberg, 2000). For summaries of task-based research, see Bygate, Skehan, and Swain (2001); Ellis (2003, 2005); Samuda and Bygate (2008); and Skehan (1998).

Based on a needs analysis, the instructor can then design explicit language instruction to enhance oral fluency. Instruction may include formulaic sequences to increase mean length of run and discourse markers to provide online planning time and reduce the length and frequency of silent pauses. Additional features could be integrated into classroom instruction to supplement free-production tasks, which alone are unlikely to have a significant impact on oral fluency. We have provided examples of each of these features below, along with suggested tasks.

Formulaic sequences

A formulaic sequence is defined as a sequence, continuous or discontinuous, of words or other elements, which is, or appears to be, prefabricated; it is stored and retrieved whole from memory at the time of use, rather than being subject to generation or analysis by the language grammar (Wray, 2002, p. 9). Formulaic sequences encompass collocations (e.g., bread and butter); phrasal verbs (e.g., run into); idioms (e.g., a piece of cake); and figures of speech (e.g., as cold as ice); as well as lexical bundles, which include:

Disappearing text

The 'disappearing text' challenge (adapted from Nation & Newton, 2009) is a form of oral reproduction and can be designed to promote automatization of selected formulaic sequences.

Disappearing Text

The instructor chooses a text of about 60 words that is relevant to learners' needs and interests and contains numerous target formulaic sequences (e.g., collocations, phrasal verbs, idioms). The instructor displays the text on the board and reads it out loud. One or two students are then asked to read it out loud. Next, the instructor deletes some of the formulaic sequences and asks another learner to read the text out loud and fill in the missing formulaic sequences. Phrases are gradually deleted, other learners read, and so on, until all the words have been deleted and the learners are repeating the passage and the formulaic sequences from memory.

Gambits

Gambits may include single words and non-lexical expressions (e.g., um), in addition to phrases (Wray, 2002), and they are defined as:

formulaic expressions whose primary role is strategic rather than propositional in nature; they serve to guide the hearer through the discourse by semantically framing propositional information (e.g., The main point is), by facilitating turn exchanges (e.g., May I interrupt for a moment?), and by marking discourse boundaries (e.g., That's all I have to say about that).

(DuFon, 1995, cited in Wray, 2002, p. 177)

The following sample task, adapted from Keller and Warner (2005), provides practice in the automatization of gambits.

Gambits Role Plays: When things aren't quite what they seem . . .

The instructor provides a sample dialogue containing gambits such as the following:

• The way things seem

• Lots of people think . . .

• It may seem . . .

• It looks like . . .

• Supposedly, . . .

• The way they are

• The truth is, . . .

• In reality, however, . . .

• But it's really . . .

• Actually, . . .

Following a focus on the meaning, use, and production of the gambits, the instructor asks learners to engage in a series of role plays such as the one below, using target expressions from the lists. With each successive production, the gambits become more automatized.

Scenario 1: Your friend is interested in buying a used car from a classmate for $2,000. Your friend doesn't have much money. Convince him/her that it would be a bad idea to buy the car.

Figure 1.

Communication strategy task.

Communication strategies

Communication strategies can be used to overcome learners' performance problems and to provide online planning time (Dörnyei & Scott, 1997). Instructors could teach achievement strategies such as circumlocution, which can be used to describe ideas for which the lexis is unfamiliar and thus assist in avoiding communication breakdown. Activities such as the communicative crossword puzzle, depicted in Figure 1 (adapted from Woodeson, 1982), require learners to formulate paraphrases or definitions to assist their partners in filling gaps in the puzzle. In so doing, learners are encouraged to repeat and develop automaticity in using such sentence stems as 'It's a tool/machine that . . .'; 'It's a person who . . .'; 'It's a place where . . .'; 'It's a type of . . .,' and so on. This activity can also be designed to reinforce new vocabulary items (using free online puzzle creation programs).

Discourse markers

Macro-markers

Macro-markers (e.g., first, on the other hand, for example, in conclusion) are used primarily to show connections between the meaning and the structure of transactional spoken discourse (Nattinger & DeCarrico, 1992). In addition to enhancing fluency, these devices facilitate comprehension (Tyler, 1992) and aid in the retention of discourse content (Chaudron & Richards, 1986). The following task, adapted from Matthews (1994), illustrates how these devices can be integrated into the ESL classroom to promote oral fluency.

Organizing Your Ideas

The instructor asks small groups of learners to prepare a brief presentation on a specific topic. For example, a group of English for Academic Purposes students could be provided with this outline:

A university education is beneficial because it

1. a. provides you with critical thinking skills;

2. b. allows you to meet people with similar interests;

3. c. can lead to a more interesting and financially rewarding career.

Students work together to develop the given topic with examples, explanations, etc., using discourse markers to provide a coherent and logical flow to the specific points being made. Such markers may include first, to begin with, second, third, next, finally, for example, in other words, in addition, moreover, therefore, in conclusion, in summary, etc. Groups may be given different topics to develop. When prepared, they can then be paired with a series of classmates from different groups to make their presentations.

Micro-markers

Micro-markers (e.g., so, well, yeah) are commonly used as lexical pause fillers in social interactional discourse to provide online planning time and to promote fluency (Nattinger & DeCarrico, 1992). The following sample activity was adapted from Dörnyei and Thurrell (1991) for that purpose.

Conversation Expansion

The instructor provides learners with a dialogue consisting of single word utterances that convey a coherent message. Working in pairs, learners alternate to construct their own dialogues of single-word utterances.

A:

Movie?

B:

When?

A:

Friday.

B:

Okay.

Learners are asked to practise the skeleton dialogues in their original form. Then they work in pairs to expand the dialogues as fully as possible without altering the basic meaning or structure. This will conceivably lead to extensive use of micro-markers:

A:

So, do you want to go to a movie or something?

B:

Well, I'm not sure. When were you thinking of going?

A:

Oh, I don't know. Maybe Friday night.

B:

Yeah, okay. Actually, that sounds like fun.

Techniques for developing fluency

Formulaic sequences, discourse markers, and other linguistic features that enhance oral fluency can be integrated into instruction (beyond use in free-production activities) by means of consciousness-raising, rehearsal or repetition, pre-task planning, and the imposition of time constraints on production. Guidelines for each of these techniques are presented in the following section. Where applicable, topics should be selected with care to ensure that they are relevant to the course content and to the needs and interests of the learners.

Consciousness-raising

Schmidt's (2001) 'noticing hypothesis' holds that conscious attention to features of L2 input is key to learning. Instructors can raise learners' awareness of factors that influence perceptions of fluency (e.g., formulaic sequences, discourse markers, pauses, fillers) by assigning form-focused tasks that include an explicit explanation of the meaning and use of these devices. As each feature is addressed, it could be added to a checklist that learners use to assess and monitor their own productions on an ongoing basis. The following consciousness-raising task is adapted from Riggenbach (1999).

Consciousness-raising: Online planning strategies

The instructor explains to learners that the ability to organize ideas in conversation is limited by time. Learners are asked which strategies they use in their native language to gain planning time. Do they use long silent pauses? Do they fill them with sounds such as um, uh, and er? Do they use particular expressions (e.g., I think . . . or What I mean is . . .) to give themselves more time to prepare their thoughts? Have they noticed similar strategies being used by native speakers of English? What sorts of fillers do native speakers use?

Learners could also be encouraged to tape (with the permission of the speakers) target language speech for later transcription and analysis in the classroom, as illustrated below (adapted from Riggenbach, 1999).

Data collection and analysis: Interactional fillers

The instructor asks learners to audio- or videotape two native speakers in a one-minute conversation with each other on a given topic. Next, in groups, the learners transcribe the conversations and identify any lexical (e.g., well, like) or non-lexical (e.g., um, er) fillers that the native speakers may have used. Then they compile the results, present their findings to the class, and discuss which fillers were commonly used and which may be more limited to specific speakers. Finally, the learners record their own conversation with a partner, compare their use of fillers with the fillers used by the native speakers in the previous exercise, and report the similarities and differences that they notice.

A related task (adapted from Riggenbach, 1999) could focus on a broader range of factors that affect perceptions of oral fluency.

Data analysis: Identifying factors that affect fluency

The instructor initiates a discussion of oral fluency, asking questions such as the following: How does fluency affect interaction? What are some of the difficulties that arise when interacting with a dysfluent speaker? What are the major factors that enhance fluency? Which are most likely to make a difference?

The instructor then introduces excerpts from television, radio, and/or film for analysis. Listening to and later reading verbatim transcripts from these samples will help learners to identify a range of features used by native speakers to facilitate fluency or give an impression of fluency. As a follow-up activity, the instructor may focus on some of the formulaic sequences, gambits, and discourse markers found in the transcripts.

Once learners have developed a heightened awareness of factors that affect oral fluency and have received explicit instruction in strategies for improving it, they should be given ample opportunity for focused, contextualized practice. Instructors can design tasks for learners based on the fluency features highlighted in consciousness-raising activities. Ongoing consciousness-raising and reinforcement, along with the reiteration of instruction, promote fluency. Above all, learners should make an effort to transfer these developing skills to new contexts. Teachers or counsellors in an ESL program could arrange contact activities with English speakers outside the classroom. For example, carefully chosen volunteer placements could be provided to give learners further speaking opportunities; in addition to enhancing fluency, these types of activities offer learners non-linguistic benefits such as the development of social networks and insights into the target culture (Dudley, 2007). Face-to-face or telephone interaction with English speakers could also be arranged; for example, advanced learners of English who aspire to attend colleges or universities in the community could be paired with students in those contexts for focused information exchanges on a weekly basis.

Rehearsal/repetition

As familiarity with language increases through repetition, fluency also improves (e.g., Bygate, 2001; Gatbonton & Segalowitz, 2005; Lynch & Maclean, 2001). Lynch and Maclean's 'poster carousel' can be adapted for learners at varying proficiency levels.

Poster presentations

Each pair of learners in a class selects and reads or listens to a different passage on a topic of interest (e.g., health issues, the environment), then prepares and displays a poster based on the text. One learner of each pair (the 'host') stands next to the poster, prepared to answer questions about the subject, while the other (the 'visitor') moves in a clockwise direction, pausing at each of the posters of other classmates for several minutes to ask questions about their topics. When the visitor returns to his or her original poster, the partners switch roles. The variety of reading/listening passages provided keeps interest high and helps to extend learners' knowledge in a given subject area.

Lynch and Maclean noted that, with repetition, responses to classmates' questions about their posters became more fluent, and learners' grammar, phonology, and lexical access and selection were also enhanced.

The next activity is based on a task type in the Test of Spoken English (Educational Testing Service, 2008). As learners provide multiple responses to the same questions, the fluency of their productions improves.

Best response

The graph depicted in Figure 2 shows Sebastian's monthly spending habits based on a $2400 salary.

Take 15 seconds to look at the graph (Figure 2 below).

In the language lab, the instructor asks students to record three or more responses to the following questions:

1. 1. Tell me about the information provided in the graph. (60 seconds)

2. 2. Sebastian would like to begin saving $200 per month for his retirement. What changes to his monthly spending would help him reach this goal? (60 seconds)

Figure 2.

How Sebastian spends his #2400/month salary.

The instructor then asks learners to carefully select and submit for assessment their most fluent response to each question.

In addition to in-class tasks, activities such as surveys can provide repetition outside the classroom. Because conducting structured surveys requires learners to interact with English speakers in their communities, the activity can enhance their fluency and confidence in speaking English.

For or against?

The instructor and/or learners choose a topic of current interest in the local media (e.g., Should the government ban the use of cell phones by drivers?). With the instructor's guidance, students formulate a series of questions related to the issue (including open-ended questions), along with openers (Excuse me, . . .) and closers (Thanks for your time). Learners practise asking and answering the questions among themselves in class. For homework, they gather responses to the survey questions from six English speakers in the community. Results are tallied, reported, and analyzed in the following class.

Tracking or shadowing

Another form of repetition is tracking or shadowing to raise awareness of thought groups and pausing and to improve automatization. This imitation technique is common in the teaching of pronunciation (e.g., Celce-Murcia, Brinton, & Goodwin, 1996). When tracking, learners repeat or read out loud the transcript of an aural passage at the same time as they hear it; this process is repeated multiple times. When shadowing, learners repeat the passage in the same way, but slightly after, rather than simultaneously with, the speaker. The following example is adapted from Nolasco and Arthur (1987).

The instructor plays a short audio recording or podcast of a passage or interaction that presents no comprehension problems for learners. Then, as the first few sentences are replayed, the instructor guides the learners in identifying pauses, thought groups, and formulaic sequences in the printed transcript. The learners listen carefully to the entire recording two or three times more on their own, following the transcript and paying attention to pauses and thought groups throughout. When they feel confident in their ability to reproduce the passage, they read it aloud several times in unison with or slightly after the speaker, paying particular attention to the target features. They continue to work individually at their own pace until they feel confident enough to make their own recordings of the passage for self-, peer, or instructor assessment.

Pre-task planning

Avery easy adaptation of a communicative language lesson is the provision of pre-task planning (Foster & Skehan, 1996, 1999; Mehnert, 1998; Skehan & Foster, 2005; Tavakoli & Skehan, 2005; Yuan & Ellis, 2003). Strategic pre-task planning may be unguided (with no specific focus); or guided, with particular attention to the content, organization, or language required to complete the task. Strategic pre-task planning time of up to 10 minutes has been shown to enhance the fluency of learners' oral productions (Mehnert, 1998; Skehan & Foster, 2005).

Time constraints

Time constraints can be imposed on productions using Nation's (1989) 4/3/2 technique, as described below.

The 4/3/2 Technique

A learner is given several minutes to mentally prepare a talk on a topic without making notes. Topics can relate to course content or to issues of wider interest. The learner is paired with a classmate and asked to deliver the talk in four minutes, while his or her partner listens. The speaker is then paired with a different classmate and given only three minutes to deliver the same talk. Finally, the speaker is paired with a third classmate and instructed to deliver the talk in only two minutes.

Nation's study demonstrated that, as learners delivered the same talk to a series of listeners under increasing time constraints, their rate of speaking increased over time, and the number of false starts, hesitations, and repeated words decreased. With each repetition, the learners gained both the confidence and, by extension, easier access to the language required to fulfil the task.

Conclusion

Because the primary goal of most communicative L2 programs is to foster communicative competence, oral fluency is an important outcome criterion. However, the development of fluency is neglected in many ESL classrooms and in the learner texts that often form the basis of communicative second language instruction. The survey of learner and teacher resources reported above shows that very few texts include a full range of activities for enhancing oral fluency. Rehearsal and repetition, consciousness-raising, and the use of discourse markers in particular are under-represented in the texts; free-production activities alone are insufficient for enhancing oral fluency.

Although research on oral fluency development has generated numerous tasks that could be used in ESL classrooms, some instructors may be unaware of them or may not incorporate them into instruction. The fluency development activities illustrated in this paper can easily be adjusted to the proficiency level of a given set of learners. Furthermore, at the discretion of ESL instructors, the activities can be incorporated effectively into various phases of traditional Presentation-Practice-Production, task-based, or other lesson frameworks to compensate for deficiencies in instructional curricula or materials. Instructors can then investigate, on their own or in collaboration with researchers, the effects of the fluency instruction. Oral narratives (e.g., picture stories) could be collected near the beginning and at the end of term for low-stakes assessment, and more importantly, for evaluation of instructional efficacy; if learners are familiar with this type of task before the pre-instruction session, differences in measurements of syllables per second over time should be attributable to instruction. This particular measurement is something an instructor could reasonably calculate over a short sample of the learners' speech, without the labour-intensive requirements of formal research. Although rating scales are commonly employed in standardized proficiency tests to assess oral fluency, learner improvement is unlikely to be reflected in such measures over the duration of a single ESL course. It is recommended that teachers use group, as opposedto individual, averages to assess the overall efficacy of fluency instruction because specific individuals may performbetter orworse for unrelated reasons (e.g., tiredness on the day of the test).

Some ESL programs may emphasize reading and writing skills, placing less focus on oral language, with the assumption that learners will develop oral fluency outside the classroom; this expectation, however, may be unfounded (Derwing, Munro, & Thomson, 2008). The integration of explicit, focused fluency activities such as those described above will enhance L2 communicative competence, thereby increasing learners' opportunities for further interaction with English speakers in the community.

Finally, we are encouraged by the growing number of formal studies of L2 oral fluency development in the L2 acquisition literature, but we see the need for collaborative work involving both classroom teachers and researchers to investigate this phenomenon further. Teachers' challenges in implementing these types of activities, determination of maintenance or attrition of fluency skills in delayed post-tests, and identification of ideal combinations of strategies for optimal results are all questions that should be addressed.