Literature Review

There are still a lot of researches to do concerning how linguistic complexity influences fluency in oral performance since previous studies related to fluency mainly focus on the discussion of definition and the comparative studies between accuracy and fluency in second language acquisition. Therefore, we have this literature review in four parts: Firstly, the classification and definition of fluency; secondly, Krashen#8217;s input hypothesis; thirdly, the factors that influence fluency; finally, the empirical studies of fluency.

As to the definition of fluency, scholars have different opinions. Though definitions are varied, they mainly lie in two groups: In a broad sense or in a narrow sense. Definitions in a broad sense deal more with EFL learners#8217; overall language using abilities regardless of qualities of their output. Like Brumfit (1984:42), who believes that fluency is ”the maximally effective operation of the language system so far acquired by the student”. Lennoch (1990) points out that high-order fluency (fluency in a broad sense) concerns more about oral language in general. When Leeson (1990; 399) distinguishes oral fluency in the narrow and broad sense, he defines the latter as ”native like rapidity”, which also emphasizes the natural flow of one#8217;s output. However, definitions in a narrow sense are closely related to EFL learners#8217; language proficiency. Lennoch holds the opinion that the low-order fluency (fluency in a narrow sense) deals more with the speed and smooth in spoken language. Leeson (1990; 399) also defines fluency in narrow sense as ”a cover term for oral proficiency”. Peter Skehan (1996) insists that fluency means language learners are able to produce speech with no hesitation or pausing. Fillmore (1979) defines fluency by giving four indexes on this issue: to have few pause, to utter sentences consistently, to communicate appropriately under any circumstance, and to use language with creativity and imagination. As to the classification of fluency, Faerch et al (1984) divides it into three types: sounding fluency, semantic fluency and syntax-lexical fluency. Faerch et al#8217;s way of classification is the most famous one and has a great influence on scholars#8217; variable-choosing. When conducting their researches, scholars like Brumfit and Zhang choose variables mainly from these three respects: temporal fluency, linguistic fluency and performance fluency, which coincides with Faerch#8217; et al#8217;s idea of classification. In these three dimensions, sounding fluency is closely related to fluency in a broad sense. Scholars like Brumfit applies this concept mainly in teaching activities and Zhang finds out several indexes to measure it like speaking rate, articulation rate, phonation time ratio and so on. Since semantic and syntax-lexical fluency are more like fluency in a narrow sense, indexes like ratio of error-free T-units and repairs per 100 syllables are widely used in doing specific researches when studying this kind of fluency.

The theory of Krashen#8217;s input hypothesis are the underlying theory of this study. Krashen#8217;s (1982) input hypothesis consists of five parts: acquisition-learning distinction hypothesis, nature order hypothesis, monitor hypothesis, input hypothesis, and affective#8212;filter hypothesis. He emphasizes the importance of comprehensible input, which calls for an understanding of the input materials by the learners. In this study, listening comprehension is of crucial significance. According to Krashen, when EFL learners know exactly what they are listening to, their retelling task may prove to be easy. But when they are listening to a story which is obscure and vague, their retelling task may turn out to be quite overwhelming. Thus it is very important for us to figure out what these specific factors are when doing our story retelling, especially in terms of fluency.

Factors that influence oral fluency come mainly from these three types: types of the tasks, linguistic input, and EFL learners#8217; language using ability. In types of the tasks, scholars like Sun (2005), Fosteramp; Skenhan (1996), and Zhang (1999) all believe that the type of L2 practice, such as spontaneous speech or story retelling, has a great impact on EFL learners#8217; performance because it also decides whether or not EFL learners have preparations beforehand and how do they know about the task. As to the linguistic input, researches on it are quite limited. Zhang (1999) puts it in his five external effecting factors yet without any specific studies. Finally, there are a large number of scholars doing researches on EFL learners#8217; language using ability. Sajavara (1987) insists EFL learners#8217; lacking in message processing abilities results in their poor oral fluency. Wei (2002; 18) argues that ”the vital factors lie in the failure to store permanently in mind the sound of second language fast as well as its pronunciation and meaning ”. Gong (2004) does researches on EFL learners#8217; lexical abilities. Raupach (1987), Towell et al (1996), and Miuamp;Sun (2006) all hold the opinion that ”sentence builders” are closely related to fluency.

As to the empirical studies of fluency, few researches cover the study of what linguistic complexity may do to fluency. However, there are large quantities of papers and studies concerning the comparative study of accuracy and fluency in second language teaching. Those papers and studies are quite different in variables-choosing. Yin (2011; 67) and Miu (2009) all consider pausing as an important criterion in measuring fluency. Yin divides those factors into language factors and non-language factors, delves into the underlying causes of these pausing and figures out the role one#8217;s language capability is playing in fluency. She defines pausing as ”a silent break of one second or longer either within sentence or between sentences”, which makes this criterion clear and definite. In this case, Miu choose 0.3 second as her criterion. By investigating undergraduates#8217; lexical competence in the way of monthly essays and doing oral tasks such as story retelling, Gong (2004) suggests that lexical phrases may influence the fluency. Zhang (1999:220; 2001) has made great achievements in doing research concerning fluency. He comes up with twelve internal indexes from time, content to linguistic and five external effecting factors like linguistic input, types of L2 practice, amount of L2 practice, quality of L2 practice and focus of L2 practice. After doing those researches, he puts forward a new theoretical model interpreting the development of fluency in L2 learning from input to output and as well as a new evaluation system. However, some shortcomings of previous studies also appear when doing researches on fluency. Scholars are not in consistency when talking about the factors and specific criterions, just as they hold different opinions about how to quantify pausing as mentioned above. Furthermore, an overall consideration may not be provided when choosing variations, the conduct of research may be impractical and incomprehensive sometimes.