國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
4
1.2 Purpose of the Study
Due to the lack of literature reporting the effects of guided planning and pair work on young learners’ oral narrative production, the present study aims to provide insight into the issue. The purpose of this study is to determine if guided planning and pair work would improve participants’ verbal output on the aspects of fluency,
accuracy and complexity. It was suggested that the constructs of learners’ oral performance were multidimensional, and the three closely linked linguistic
components (i.e. complexity, accuracy and fluency) could comprehensively reflect the major aspects of learners’ proficiency and language progress (Ellis, 2008; Ellis &
Barkhuizen, 2005; Skehan, 1998).
1.3 Significance of the Study
There is a prevailing trend to teach conversation with both inductive attention to meaning and deductive attention to form (Brown, 2001). It is hoped that the present study may contribute to the pedagogical field where students will not only learn to speak, with a focus on meaning, but also speak to learn, with a focus on form.
Additionally, it is believed that the study will shed some light on the conflicting results in the previous task planning studies, particularly in regards to accuracy and
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
5
complexity.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
6
Chapter Two Literature Review
In this chapter, we will look at the literature on the use of planning time in five parts. The first part provides the theoretical background of task-based instruction (TBI), with a focus on speech instruction. The second part presents literature about the effects of planning time on oral production. Moreover, two solutions are proposed to resolve the conflicting results. The fourth part describes the performance
descriptors for oral assessment of language learners. Lastly, the purpose and research questions of the present study are stated.
2.1 Theoretical Background of Task-based Speech Production Research
The ability to speak in a foreign language has been a primary issue in language teaching after the prosperity of CLT (Nakamura, 1993). Learners are encouraged to take an active role in using the language in order to obtain communicative
competence (Brown, 2007). Therefore, considerable attention has been given to students’ speaking opportunities, which used to be neglected in grammar translation method. TBI has been proposed as a valuable technique for the development of oral communication ability within the CLT approach (Nunan, 2004).
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
7
TBI has been well-received and widely discussed in language teaching and research since 1970s (Ellis, 2003; Long & Crookes, 1992; Shehadeh, 2005). With the surging popularity of TBI, various definitions of tasks have been proposed by
researchers. Some of the definitions are very broad (e.g., Carroll, 1993; Long, 1985) while others narrowed the meaning of tasks to more specific activities in the field of pedagogy (e.g., Breen, 1987; Willis, 1996). For example, Long provides a very general and non-linguistic definition for tasks: a piece of work people voluntarily perform or an obligation for performance in daily life. Breen, however, provides a more restricted definition: activities carried out in a language classroom, with a goal in real life. In this study, the definition of Nunan (1989) and Skehan (1998) best suit the purpose of this study. Tasks are considered as classroom activities where learners pay attention to the message they want to convey or the message they need to receive from others. These classroom tasks resemble those learners use their first language to conduct in daily communication. For example, telling stories, ordering food, greeting people, and as such, are very common tasks in L2 language teaching. This definition agrees with the activities described as tasks in the literature which was reviewed and in this study which was conducted.
There are several reasons for TBI to gain popularity in the field of language teaching. First, TBI is more student-centered since the materials are connected with
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
8
students’ personal experiences. The authentic materials in TBI can retain students’
focus on the target language input through meaningful communication (Ellis, 2001;
Swain, 1995). Moreover, the principles for content selection in TBI are more concrete and are based on practical needs (Bygate, 1999). Therefore, the evaluation is
outcome-based (Nunan, 2004). Students are considered qualified because they have the ability to accomplish an authentic and relevant task, not because they are able to complete a contextually detached paper-pencil test. Finally, TBI opens up the
possibility for instructors to direct participants’ attention to specific linguistic aspects by deliberately manipulating task design and implementation situations. Thus, learners stand a better chance to sharpen specific language skills (Candlin, 1987;
Samuda, 2001; Skehan, 1998).
Looking at TBI from the psycholinguistic perspective, Long’s (1996) interaction hypothesis is perhaps one of the most fundamental and essential rationales for the method. He posits that authentic conversation triggers learners’ self-realization. The mutual exchange can facilitate learners’ target language acquisition, particularly when it comes to the negotiation of meaning and modifications to the target language.
However, there is some criticism on interactions in TBI. When conducting tasks, learners tend to communicate for pragmatic purposes by paying focal attention to meaning and minor attention to forms (Ellis, 2003; Swan, 2005). TBI is also criticized
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
9
for the likelihood that students will focus too much on completing the task rather than practicing language (Richards, Jack, & Nunan, 1990). Interaction in TBI might facilitate comprehension about meaning, but this does not necessarily lead to internalization of L2 forms (Ellis, 1991).
Levelt’s (1989) speaking model might further explain why learners tend to prioritize the language aspect of meaning over the aspect of grammar in task-based interactions. The production model is subdivided into three processing components:
the conceptualizer, the formulator, and the articulator. Before the production of speech, the conceptualizer operates to work out preverbal messages, which is composed of basic semantic categories (e.g., persons, actions, and places). The semantic
representation, serving as basis for further linguistic expression, is then transmitted to the formulator. The formulator is responsible for grammatical and lexical encoding. In this formulation process, appropriate words and correct grammar are selected for the concept message, followed by a production of phonologic plan. At the last stage, the phonologic plan is transformed into verbal output by the articulator. Levelts’ speaking model indicates that, under ideal conditions, the conceptualizer operates first and continuously feeds pre-linguistic message into the formulator and the articulator during the whole process of speech production. For native speakers, the three stages of processing (i.e. idea generation, grammatical and lexical modification, and speech
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
10
articulation) can work well simultaneously due to an abundant storage of
prefabricated language chunks (Pawley & Syder, 1983). However, such automatic language processing is hard to achieve for language learners, particularly those of beginning proficiency (Skehan, 1998). Language learners might speak slowly or even pause when the conceptualizer, formulator and articulator compete for limited
attentional resources. It appears that the conceptualizer, responsible for meaning processing, holds obvious superiority over the formulator and articulator in competition for the finite cognitive capability (Van Patten, 1990).
In accordance with the risky tendency for meaning instead of form in TBI (Skehan & Foster, 2001), judicious manipulation is called for balancing learners’
attention on form as well as on meaning. For instance, some researchers have
suggested that pre-task planning time (i.e. the offer of time for learners to manipulate linguistic device before conducting tasks) might decrease the burden of processing the conceptual message (e.g., Bygate, 2001; Ellis, 1987; Mehnert, 1998). It is assumed that as long as the conceptualizer has accomplished part of the message formulation during planning time, there will be more resources left for the formulator and the articulator to encode grammatical information and transform spoken utterance respectively. Accordingly, learners can spend more efforts to attend to grammatical accuracy. Other researchers have proposed form-focused instruction to facilitate
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
11
grammatical learning in a meaningful context by implementing pedagogical interventions (e.g., Doughty & Williams, 1998). It is suggested that the extent to which the task requires grammatical accuracy can affect the efficiency of language acquisition (Loschky & Bley-Vroman, 1993). Therefore, if grammatically correct languages are involved in task design as essential features to successfully complete a task, language acquisition can be more effective (Loschky, 1994). Briefly, form-focus elements have been proposed to be integrated into the framework of TBI (Willis, 1996), in the hope of directing learners’ attention to forms (Skehan, 1998).
2.2 Utilization of Planning Time to Balance Communication and Grammar
Theoretically motivated by the notion of focus on form, pre-task planning has been adopted to develop both grammatical and communicative ability (Ellis, 2005b).
Ellis (2005a) has divided planning into pre-task planning (i.e. the provision of time for learners to manipulate linguistic device before performing tasks) and within-task planning (i.e. the planning that occurs simultaneously when learners are performing tasks). In the following, the researcher focuses mainly on pre-task planning since all the literature mentioned in this study and the planning performed in this study pertain to only pre-task planning. Pre-task planning can take two forms. It can be rehearsal (i.e. the opportunity for learners to practice the whole task once before they actually
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
12
carry out the task for a second time) or strategic planning (i.e. the opportunity for learners to make a draft version of content and plan for the language usage, without the opportunity to rehearse the whole task).
Several theories have been put forward for the beneficial role of pre-task
planning on enhancing learners’ development of interlanguage. Ellis (1987) suggested that planning could provide a better opportunity for planners to access grammatical concepts and put them into practice. According to Ellis, the internalization and spontaneous processing of grammar is not easy for L2 learners. He suggested that through planning procedure, learners would be more likely to subconsciously receive, absorb and utilize grammar rules in the production. Furthermore, Crookes (1989) has indicated that planning can contribute to interlanguage development by increasing the complexity in learners’ use of L2. Based on Crooke’s mental message-processing hypothesis, planning is described as a manipulable context in TBI. Students could bring their interlanguage capability into full play, produce more advanced speech, and acquire language skills under a maneuvered planning situation. Later, Skehan (1998) reconceptualized the role of planning in TBI as a coordinator to regulate mental burden, instead of a director to allocate attention to either one of the three aspects (i.e.
complexity, accuracy and fluency). It was advocated that planning before tasks could elude the danger of dedicating attention only to meaning by alleviating
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
13
communicative pressure. In turn, the liberation of moderate amount of learners’
memory space could be utilized to improve other linguistic elements, such as form correctness. Based on Skehan’s rationale, the incorporation of planning into TBI would judiciously distribute learners’ attention among fluency, complexity, and accuracy, leading to a balanced boost of speech output in terms of the three dimensions.
Motivated by the above rationales, a substantial amount of empirical research have been carried out to test the assumption that pre-task planning could assist learners’ inter-language in terms of accuracy, complexity and fluency. First, previous literature predicted that pre-task planning would lead to more grammatically correct expressions since extra cognitive resource was released to supervise grammatical accuracy during planning phase (Ellis, 1987). Nevertheless, several studies have invalidated such prediction (Wendel, 1997; Yuan & Ellis, 2003). Second, it was found that planning played an advantageous role in triggering speakers’ utterance
complexity (Crookes, 1989). Crookes suggested that speakers who faced less time pressure under planning condition could take ultimate risks to draw the more complex syntax and lexicon from inter-language. Such finding has been opposed by some later studies (Elder & Iwashita, 2005; Mehnert, 1998; Mochizuki & Ortega, 2008; Tajima, 2003). Moreover, it was assumed that planners paused less and spoke more fluently
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
14
than non-planners since the cognitive load of planners was lighter (Fathman, 1980).
Fathman found that planners’ cognitive burden was alleviated since they did not have to attend to the formulation of utterance content and the production of speech
simultaneously. The supposition has been confirmed by most researchers (Foster, 1996; Kawauchi, 2005; Mochizuki & Ortega, 2008; Skehan & Foster, 2005).
As noted above, the effects of pre-task planning on learners’ oral production have been somewhat contradictory in the previous research, particularly on the aspects of accuracy and complexity (Ellis, 2009; Ortega, 1999). In an attempt to solve the discrepant results of pre-task planning on accuracy and complexity, two possible solutions, guided planning and pair work, are presented. An elaboration of conflicting results of previous research is succeeded by a recommendation of possible solution and its related rationale background. It is believed that a review of previous research in pre-task planning helps to provide some perspectives for further exploration into such mixed results.
2.3 The Role of Guided Planning in Pre-task Planning 2.3.1 Discrepant Effects of Pre-task Planning on Accuracy
On the whole, the issue of whether pre-task planning could improve the grammatical component of language has been debated in the research (Ellis, 2009;
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
15
Ortega, 1999). Some researchers have viewed pre-task planning as a valuable technique in improving grammatical accuracy (e.g., Mochizuki & Ortega, 2008;
Sangarun, 2005), whereas others have argued to the contrary (Crookes, 1989; Wendel, 1997; Yuan & Ellis, 2003). The various assumptions of pre-task planning noted above are exemplified by the following statements.
It was maintained that learners did not tend to process form and meaning individually but rather ―engaged in solving form-in-meaning problems‖ in regard of planning operationalization (Ortega, 2005, p. 106). Consequently, learners would not change their mind on contributing focal attention to meaning unless there were very specific instructions engaging learners to focus on form. Another neutral point of view was that it was difficult for learners to give consideration to all three language facets due to the processing capacity trade-off (Skehan, 1996). In other words, when learners intentionally attended to accuracy, complexity and fluency would suffer, and vice versa. Yet another assumption was offered by Ellis’s (1987) study, which held a positive attitude towards the enhancement of productive accuracy through pre-task planning. It was proposed that planning could provide space for learners to
subconsciously devote their attention on formal aspects of the language, understand and acquire grammar rules. Subsequently, learners would automatically process these grammar points which would be used as a resource to utter grammatically accurate
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
16
speech.
2.3.2 Possible Solution: Guided Planning
To solve the problem of conflicting results of pre-task planning on accuracy, Ortega (1999) proposed the idea of leading learners to focus on the formal and systematic features of the language. Such guidance to manipulate learners’ attention during planning time was termed as guided planning (Sangarun, 2005). It was defined as a kind of pre-task planning, which directed learners’ attention to the meaning and/or grammar of their output, while in unguided condition, learners received no specific instruction. There are some advantages for incorporating guided planning in TBI. First, guided planning could ease learners’ communicative burden as well as release cognitive space (Lennon, 1989). In this way, the freed-up attentional resource may support learners to prioritize and to attend to specific task requirements (Ortega, 1999). For instance, Ortega suggested the use of a grammar-centered instruction worksheet in TBI to redirect learners’ focal attention to language forms in the context of meaningful communication. Specifically, when the use of accurate grammar became an essential requirement of tasks through the inclusion of form-focused worksheets, learners’ accuracy of language would be enhanced in the natural and meaningful task context (Loschky & Bley-Vroman, 1993).
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
17
Inspired by the above rationales, Mochizuki and Ortega’s (2008) study was noteworthy for its significant findings that guided planners’ planned output was significantly more accurate. In their study, the guided planners produced considerably more accurate relative clauses than both the unguided planners and participants who did not plan at all. Hence, optimal guidance prior to the task may facilitate learners in planning effectively, especially on the grammatical accuracy.
As indicated previously, while substantial studies have been done on the impact of pre-task planning on the accuracy of learners’ oral output, no conclusive finding is yet available. Even though there is a strong assumption of a link between guided worksheets and oral accuracy performance, scant empirical evidence has been found to establish a direct relationship between the two variables. Mochizuki and Ortega’s (2008) study was the only one that has indicated the value of guided worksheets to maximize the effectiveness of planning time in promoting grammatical accuracy.
Obviously, more empirical investigations are needed to gather evidence in support of such an assumption. Furthermore, most pre-task planning studies have centered on the participants of adults or university students who were generally more mature
cognitively and meta-cognitively (e.g., Wendel, 1997; Mehnert, 1998; Kawauchi, 2005), whereas the exploration into young students or language beginners has been neglected (Mochizuki & Ortega, 2008). As has been discussed, it is prudent to
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
18
investigate deeper into the integration of guided planning in TBI, with a focus on young or beginning English learners.
2.4 The Role of Pair Work in Pre-task Planning
2.4.1 Discrepant Effects of Pre-task Planning on Complexity
There have been conflicting results about the effects of pre-task planning on the aspect of language complexity in the previous studies (Ellis, 2009; Ortega, 1999).
Some researchers have found no effects of pre-task planning on the complexity aspect (Elder & Iwashita, 2005; Gilabert, 2007; Mehnert, 1998; Mochizuki & Ortega, 2008;
Wigglesworth, 1997), whereas others have reported significant results (Bygate, 1996;
Guara-Tavares, 2008; Skehan & Foster, 2005; Yuan & Ellis, 2003).
It was pointed out that learners’ proficiency is of decisive importance when considering the effects of planning on complexity performance (Kawauchi, 2005;
Mochizuki & Ortega, 2008; Ortega, 1999). Kawauchi claimed that if students were advanced language learners, they might not benefit much from the opportunity of planning because of a ceiling effect. He also believed that since advanced learners could spontaneously and efficiently draw upon their abundant L2 knowledge, they seemed to handle tasks quite well with no planning time provided. Therefore,
planning was assumed to be more helpful for less advanced learners. In contrast to the
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
19
conservative attitude towards the effect of planning on speech complexity, an
optimistic opinion was held by Crookes (1989). He proposed that learners, whether of advanced or beginning proficiency, would be motivated to make the best of their upper level of interlanguage while planning without so much time pressure, thereby producing more complex language.
While the focus of most task planning studies has been on cognitive aspects, such as how to judiciously manipulate limited attentional capacities, the neglect of the context variable has also been proposed as a possible reason resulting in such
inconsistent findings (Batstone, 2005). Batstone suggested that the effects of social interaction would not be less important than cognitive processing.
2.4.2 Possible Solution: Pair Work
Following the call for more of a focus on context, there was a trend in research to involve pair work within oral assessment in both first and second language classrooms
Following the call for more of a focus on context, there was a trend in research to involve pair work within oral assessment in both first and second language classrooms