Task-Based Language Teaching proposes the notion of task as a central unit of planning and teaching. There is now a considerable volume of literature on task （Prabhu， 1987； Nunan， 1989； Wills， 1996； Eills， 2003）. In addition， task has increasingly been used as a unit for research into second language acquisition （Crookes &Gass， 1993）. Therefore， task has become an important tool for the language teacher and a basic unit for language research. The concept of task itself is not a new one. In second language pedagogy， different scholars have put forward different interpretations from different aspects. As Willis （1996： 32） emphasizes that the target language should be used for the communicative goal to show their achievement of language learning in a task. Nunan （1989： 8） also indicates that learner should interact with others in the target language， but he emphasizes that the use of the target language should focus on meaning rather than form. However， （Wang， 2006） suggests that task should be both form-focused and communication-focused. Although definitions of task vary in TBLT， there is a commonsensical understanding that task is an activity or goal that is carried out the target language. From a psycholinguistic perspective， a learners language system develops through communicating in the target language （Marion & Robert， 2008）， in other words， individuals acquire a foreign language through the process of interacting， negotiating and conveying meanings in the language in purposeful situations.Thus， the application of task in the classroom provides a better context for the activation of learning processes than form-focused activities， and hence maximized the opportunities for language learning. Language learning depends on immersing students not merely in “comprehensible input” but also in tasks that require them to negotiate meaning and engage in naturalistic and meaningful communication（cited in Richards & Rodgers. 2008）.

Krashen （1981； 1982） has long insisted that comprehensible input is the one necessary criterion for successful language acquisition. His Input Hypothesis claims that the ability to speak fluently in a second language will come on its own with time （Lightbown& Spada， 2002； Jahnson， 2004）， that is， speaking fluency is not taught directly， speaking ability merged after the learner has built up competence through comprehending input. The Input Hypothesis also proposes that a learner can move from stage i to a stage i+1 by understanding language containing i+1 （Eills， 1999； Larsen-Freeman， 2000）. Thus， students should be allowed time for grammar and vocabulary learning so that they have sufficient linguistic competence， these input can be structure-based with a focus on forms， and task， particularly the cognitively challenging one， can provide sufficient opportunities for students to develop fluency. Thus， task plays an important role in developing learners accuracy and fluency. Skehan （1998： 97） suggests that in selecting or designing task there is a trade-off between cognitive processing and focus on form. More difficult， cognitively demanding task reduce the amount of attention the learner can give to the formal features of messages， something that is thought to be necessary for accuracy and grammatical development （cited in Richards & Rodgers. 2008）， in other words， if task is too difficult， fluency may develop at the expense of accuracy， only the balance of accuracy and fluency can ultimately promote students learning. The grading of tasks is particularly complex issue because of the many different elements that contribute to task difficulty， all of which overlap and influence each other， and some scholars provide their analysis of factors accuracy and fluency， in this study， a taxonomy of factors offered by Candlin （1987） will be used. This study is therefore carried out to explore the impact of task difficulty on learners accuracy and fluency and the method of adjusting task difficulty to balance accuracy and fluency.

Reference：

[1]Marion. & Robert L. 2000. The Places of Tasks in the Language Classroom. In Psychology for Language Teachers. 223-228. Beijing： Foreign Language Teaching and Research Press and Cambridge University Press.

[2]Richards， J. C. & T. Rodgers. 2000. Approaches and Methods in Language Teaching. Beijing： Foreign Language Teaching Press.

[3]Mitchell， R. & Florence， M. 1998. Second Language Learning Theories. London： Arnold.

[4]Ellis， R. 2003. Task-based Language Learning and Teaching. Oxford： Oxford University Press.

[5]Skehan， P. 1996. Second Language Acquisition Research and Task-based Instruction. In Willis.J. & D. Willis（eds）， Challenge and Change in Language Teaching， 17-30. Oxford： Macmillan Heinemann.

作者簡(jiǎn)介：張嬌，女，民族：漢，籍貫：四川成都，碩士在讀，單位：四川師范大學(xué)外國(guó)語(yǔ)學(xué)院，研究方向：學(xué)科教學(xué)（英語(yǔ)）。