CHAPTER ONE INTRODUCTION
This chapter commences with the background of the research together with the rationale of the present study in section 1. Several key terms used in the study are defined and elucidated for further clarification, including implicit knowledge, explicit knowledge, metalinguistic knowledge, self-perceived proficiency, self-regulation and language achievement. Finally, the significance of the study and the overall organization of the thesis are stated in the final section of this chapter.
1.1 Background
Since the late 1970s, the development of the Communicative Approach has engendered the inception of Communicative Language Teaching, which advocates fluency and meaning, as opposition to language accuracy and form accentuated in Grammar Translation Method and Audio-Lingual Method. Researchers’ increasing attention to learners’ communicative competence reflects the observed fact that students’ capacity to make grammatically correct sentences does not guarantee their ability to appropriately use the language in other occasions of their daily lives (Widdowson, 1978). To improve this, Communicative Language Teaching promotes the importance of exposing learners to communicatively-oriented activities and developing their ability to use the language properly in daily communication.
One of the researchers who emphasized and advocated fluency and meaning in language learning was Krashen (1981), asserting that the sufficient provision for learners with comprehensible input within learners’ understanding or slightly beyond their present language levels (i.e. i+1) could facilitate their second language learning.
2
What Krashen described was the process in which children acquired their first language without paying conscious attention to grammar and form, or receiving any explicit instruction. Contrarily, many researchers contended that learners’ L2 performance and proficiency could not be achieved only by placing emphasis on meaning (Alderson et al., 1997; Harley & Swain, 1984; Lyster, 1987). Explicit instruction and grammar-focused activities were thus gaining support from researchers (Dekeyser, 1995; Han & Ellis, 1998; N. Ellis, 1994; R. Ellis, 2004; Terrell, 1991). Celce-Murcia (1991) and Long (1988) also stressed the importance of form-focused instruction and the necessity to incorporate grammar instruction into a meaningful context.
The pendulum swung from communicatively-oriented practices back to explicit instruction. In addition, explicit knowledge as well as implicit knowledge was obtaining attention from researchers. The dichotomy and the respective definitions of the two constructs were proposed by Ellis (2004, 2005). Moreover, the investigation about the relationship (e.g., Green & Hecht, 1992; Hu, 2002; Macrory & Stone, 2000) or the interaction (e.g., Bialystok, 1994; Dekeyser, 1998; Krashen, 1982; Paradis, 1994; Sharwood-Smith, 1981) between implicit and explicit knowledge was conducted as well.
Implicit knowledge and explicit knowledge, the two juxtaposed constructs, had distinct characteristics respectively. According to Ellis (2005), implicit knowledge is unconscious and unanalyzed while explicit knowledge involves learners’ conscious attention to form and is verbalizable. The process of the acquisition of implicit knowledge is analogous to the way children acquired their L1, without paying deliberate attention to grammar and form and can be automatically processed. On the contrary, learners’ explicit knowledge is generally developed and promoted through formal instruction. The trigger of the application of explicit knowledge requires
3
learners’ conscious attention and it involves learners’ controlled processing.
Regarding the relationship and interaction between implicit knowledge and explicit knowledge, many researchers attempted to explore the extent (if any) that implicit knowledge could possibly contribute to explicit knowledge or vice versa. It was discovered that learners’ ability to use implicit knowledge exceeded their ability to use explicit knowledge (Green & Hecht, 1992). Macrory & Stone (2000) found a disparity in learners’ use of explicit knowledge and spontaneous production, which could be regarded as a representation of implicit knowledge. According to Hu (2002), the provision of consciousness-raising tasks could enhance participants’ accuracy of structures if they paid attention to the formal aspects of language and applied their explicit knowledge. In addition to the investigation on the relationship between implicit knowledge and explicit knowledge, some researchers explored the issue of transformation or interaction between the two constructs. Dekeyser (1998) and Sharwood-Smith (1981) claimed that explicit knowledge was not only related to implicit knowledge but also could be converted into implicit knowledge through practice. However, Krashen (1982) asserted that acquisition and learning were two ways for adults to learn a language and acquisition was more important than learning.
It was impossible for learning to transform into acquisition but instead, learning could serve as a monitor for the acquisition system.
To sum up, explicit knowledge plays a decisive role in implicit knowledge and language learning. In addition to explicit knowledge, metalinguistic knowledge, which is defined as learners’ explicit knowledge about the language (Roehr, 2006) or
“the knowledge of the technical terminology needed to describe language” (Ellis, 1994, p.714), is gaining attention from researchers as well. Numerous metalinguistic studies were conducted with different ways of operationalizing and measuring metalinguistic knowledge and most studies adopted the similar research design to
4
investigate the link between metalinguistic knowledge and L2 performance. However, these studies presented inconsistent findings.
Metalinguistic knowledge, according to Alderson et al (1997), was regarded as a subsection of explicit knowledge. Sharing certain characteristics with explicit knowledge, metalinguistic knowledge was also verbalizable and its elicitation required learners’ conscious attention. A variety of instruments for measuring metalinguistic knowledge were developed. Many researchers operationalized metalinguistic knowledge as learners’ ability to identify, correct, and provide verbal explanations about L2 features (Elder and Manwaring, 2004; Green and Hecht, 1992;
Hu, 2002; Morris, 2003; Renou, 2001; Roehr, 2007; Sorace, 1985). Macrory & Stone (2000) required participants to recall the rules of a particular structure and the situations in which they would use the structure. However, verbal explanation might complicate the results due to the subjectivity issue involved. In order to avoid the subjectivity that verbal explanation may involve, Ellis (2005) developed the metalinguistic knowledge test, which includes an untimed computerized multiple-choice test with two sections. On account of the subjectivity issue involved in verbal explanation, the researcher in the present study designed another metalinguistic knowledge test with multiple choices on the basis of the first section of Ellis’s (2005) metalinguistic knowledge test, which was a more independent and objective measure and was proved to be a representative test for metalinguistic knowledge as scores on the metalinguistic knowledge test loaded on explicit knowledge. However, the metalinguistic knowledge test used in Ellis’s (2005) study only included 17 grammatical structures and was not comprehensive and representative enough to fully represent L2 learners’ metalinguistic knowledge. Hence, the researcher in the present study established an item bank of 50 items which focus on a wide range of grammatical structures that the participants have learned in junior
5
high school in the hope of measuring learners’ metalinguistic knowledge in a more comprehensive manner.
The most typical research design adopted by previous research was the investigation of the relationship between metalinguistic knowledge and L2 performance. While some studies revealed a weak relationship (e.g., Alderson et al., 1997; Elder et al., 1999; Han & Ellis, 1998), some researchers corroborated a significantly positive correlation (e.g., Morris, 2003; Roehr, 2007; Renou, 2001) between metalinguistic knowledge and L2 performance. As a result, one of the present study’s purposes is to provide more evidence to confirm the relationship between metalinguistic knowledge and L2 performance through correlation analysis and the effects of metalinguistic knowledge on L2 performance will be further explored through hierarchical regression analysis. However, L2 performance in previous studies was mostly determined from various dimensions. In Roehr’s (2007) study, learners’ L2 proficiency was determined as their competence in grammar and vocabulary in a 45-item language test. Renou (2001) measured the participants’ L2 French proficiency through a proficiency test which included listening, reading, and cloze sections. Elder & Manwaring (2004) determined learners’ learning outcomes as their performance in a Chinese achievement test. As a result, in the present study the participants’ L2 performance will not be determined exclusively as their performance in one language test that may contain items addressing different aspects of proficiency;
instead, their L2 performance will be determined as their actual performance on the first monthly exam and their general self-perceptions of their proficiency on listening, speaking, reading, and writing ability. By doing so, the present study can further investigate what factors will contribute to their own perceptions of proficiency and language achievement and analyze if self-perceived proficiency and language achievement are influenced by different sets of variables. Additionally, the effects of
6
metalinguistic knowledge on self-perceived proficiency and language achievement can also be examined.
In addition to the relationship between metalinguistic knowledge, self-perceived proficiency and language achievement, more metalinguistic studies that incorporate other kinds of factors are needed since metalinguistic knowledge is influenced by many different factors in addition to L2 performance. For example, Paradis (2004) indicated the potential influence of motivation on metalinguistic knowledge by asserting that metalinguistic knowledge, linguistic competence, pragmatics and motivation were four cerebral systems that were crucial to language learning. Obviously, metalinguistic knowledge can possibly be affected by motivation and Paradis (2004) suggests that motivation needs to be taken into account in metalinguistic studies. As seen from the aforementioned metalinguistic studies, few of them incorporated motivation into the research and considered the underlying effect of motivation on metalinguistic knowledge. Furthermore, motivation has been believed to play a decisive role in language learning but most studies have addressed the effect of motivation on vocabulary learning (Elley, 1989; Gardner & MacIntyre, 1991; Tseng & Schmitt, 2008) and learning strategies use (Biggs, 1988, 2003;
Gardner, et al., 1997; MacIntyre & Noels, 1996; Schmidt & Watanabe, 2001) with scarce studies investigating metalinguistic knowledge and motivation simultaneously except for Perales and Cenoz’s (2002) study which included attitude, motivation, anxiety and metalinguistic awareness in their research. Therefore, in addition to metalinguistic knowledge, the present study also subsumed motivation into the research design and explored the effects of motivation on self-perceived proficiency and language achievement. Metalinguistic knowledge and motivation will also be juxtaposed for comparison to see if they have different effects on self-perceived proficiency and language achievement.
7
In addition to motivation, another affective variable, self-regulation, has been found to be essential in language learning and many different theories were proposed to explain the self-regulation processes (i.e., operant, phenomenological, social-cognitive, information processing, volitional, Vygotskian, and constructivist).
However, few studies on self-regulation juxtapose self-regulation and variables such as motivation, metalinguistic knowledge and L2 performance and investigate their relationships with each other.
In conclusion, explicit knowledge, especially metalinguistic knowledge, has been gaining attention from researchers and many studies have been conducted to examine the relationship between metalinguistic knowledge and L2 performance.
Even though metalinguistic knowledge has been considered to play an indispensable role in L2 learning, its contribution to and relationship with L2 proficiency were not confirmed and still needed further empirical evidence. Hence, the present study sets out to confirm the basic relationship between metalinguistic knowledge and L2 performance by conducting correlation analysis at the first phase. Besides, since L2 performance was usually determined as performance in certain language achievement tests addressing different aspects of the target language, the present study is original in dividing L2 performance into two dimensions (i.e. self-perceived proficiency and language achievement) and investigating the effect of metalinguistic knowledge on the two dimensions. Besides, according to Paradis’s (2004) claim, metalinguistic knowledge is highly related to motivation but most motivational studies only limitedly address its effect on vocabulary and strategies use without investigating motivation and metalinguistic knowledge simultaneously except for Perales &
Cenoz’s (2002) research. Previous studies on self-regulation scarcely incorporate metalinguistic knowledge, motivation and L2 performance altogether. Consequently, the present metalinguistic study is original in incorporating affective variables (i.e.
8
motivation and self-regulation) and determining L2 performance in the two dimensions—learners’ actual language achievement and learners’ own perceptions of their proficiency to investigate their basic relationships with each other and the potential effects of metalinguistic knowledge, motivation and self-regulation on the two dimensions of L2 performance.
1.2 Definition of Terms