to produce output containing target forms, again enabling learners to notice the gap between their current TL ability and the correct use of the target form (Swain 1985, 2005).
Teacher education: the gap between SLA research and pedagogy
Regarding past and current approaches to teacher education, the point has often been made (Crandall 2000: 35) that traditional approaches have usually been top-down, viewing teachers as passive knowledge recipients, whereas current constructivist approaches emphasize active roles for teacher cognition, reflection, and research. Widdowson (1990: 62) distinguishes between teacher training and teacher education, arguing that training has traditionally been viewed as ‘a process of preparation towards the achievement of a range of outcomes which are specified in advance’, and the development of skills for predictable situations. Teacher education, however, allows for unpredictability and equips prospective teachers for ‘situations which cannot be accommodated into preconceived patterns of response but which require a reformulation of ideas and the modification of established formulae’ (ibid.: 62). This view of teacher education as the flexible development of professional knowledge to be applied when needed is emphasized in the present volume.
Two areas of SLA have been identified by R. Ellis (1997a) as having particular relevance for teacher education. One is the role of input and interaction, and the other is the role of FFI. R. Ellis (2001a) considers FFI to be any instructional activity, planned or incidental, that is used to draw the learner’s attention to language forms. A number of studies have investigated the role of FFI in SLA and their findings have been reviewed extensively (for example, Long 1983; Larsen-Freeman and Long 1991; R. Ellis 1994a, 1997a, 2001a, 2001b; Spada 1997; Doughty and Williams 1998a; Norris and Ortega 2000, 2001). However, despite these reviews, there has often been a gap between SLA research and its successful application to language pedagogy (Stenhouse 1981, 1983; Widdowson 1990; Crookes 1997a, 1997b, 1998; R. Ellis 1997a, 1997b, 2001a; Lightbown 2000; Liu and Master 2003; Nassaji 2005). For example, surveying English as a second/foreign language teachers, Nassaji (2005) found that while many teachers acknowledged the importance of SLA research, few mentioned that they regularly read such research. One reason for this could be that research results are often published in venues not easily accessible (Crookes 1997a; R. Ellis 1997a). In addition research reports are often perceived as too theoretical, with findings not directly applicable to classroom practice. Robbins (2003: 59) stated:
To date, much SLA research remains within categorical lists, little of which have been translated into teaching materials, learner expectancies, or topics in teacher training courses. To some degree this trend has resulted in theory for theory’s sake.
The purpose of the current book is to address this gap by providing chapters authored by SLA experts who are language teachers or teacher educators. The main theme is how theory and research in FFI can inform classroom pedagogy and teacher education. However, for theory and research to be useful to teachers, their connection with pedagogy should be made explicit so that they can be conceived of as relevant. For example, with respect to pedagogical grammars, defined as ‘the types of grammatical analysis and instruction designed for the needs of second language students’ (Odlin 1994: 1), it has been noted that grammar can be taught as prescription, description, an internalized system or as an axiomatic system, the latter being the kind of pedagogical grammar needed for teacher education (Fotos 2005). As the articles in a recent book on grammar teaching in teacher education emphasize (Liu and Master 2003), although teachers believe that pedagogical grammar is essential for the language classroom, many find it challenging – not only because it is difficult to learn and teach, but also partially because of the emphasis on communicative pedagogy many received during their training. Thus, the questions the authors have addressed in this volume are those asked by language teachers, and the research reported has been conducted from the vantage point of informing L2 pedagogical practices.
Theoretical research in language pedagogy
Considerable research has been conducted over the past few decades on the ways in which second languages can be best taught. R. Ellis (2001a) provides an overview of these studies outlining their origins, the research questions they have addressed, and the research methods they have used. (See also Lightbown 2000; Mitchell 2000; see Fotos 2005, for a historical survey.) As R. Ellis points out, much of the early SLA research has been method-driven with the aim of comparing language teaching methods that differed in the extent to which they taught language forms explicitly or implicitly (for example the large scale experimental research projects conducted in the 1960s and 1970s). However, these studies did not indicate that one method of teaching was superior to another. Subsequent research has become more theory-driven in focus and is now mainly conducted to test theoretical claims about second language (L2) acquisition processes (R. Ellis 2001a). These claims include those related to the role of attention and noticing in SLA (Schmidt 1993, 1995, 2001; Tomlin and Villa 1994; Leow 1998, 2001a), the relationship between instructed SLA and developmental sequences (Pienemann 1984, 1989), and the relationship between explicit and implicit knowledge and the effects of frequency on cognition and SLA (for example, Reber 1976, 1989, 1993; Bialystok 1990, 1994; N. Ellis 1993, 1994, 2002a, 2002b; DeKeyser 1995, 1997, 1998, 2005a; Robinson 1996; R. Ellis 2002a).
Although theory-based research has not been directly concerned with pedagogical issues, it has produced many new insights regarding effective L2 instruction. It has increased our understanding of the complexity of the processes underlying the learning of second languages and how they are affected by formal instruction. For example, research into the role of attention has shown that although noticing or attention to form is a crucial factor in learning an L2, it is not clear what constitutes attention and how attention functions in learning (Sharwood Smith 1981, 1993; Schmidt 1990, 1993, 1995, 2001; Tomlin and Villa 1994; Robinson 1995, 2001; Leow 1997, 2001a, 2001b, 2002; Wong 2001). Moreover, it has been shown that attention and noticing interact with the learning task and context, as well as with various cognitive processing variables (Robinson 1995, 2001, 2005; Skehan 1996b, 1998; N. Ellis 2002a, 2002b). Findings from studies on developmental sequences have indicated that, although instruction may have facilitative effects on SLA, its effectiveness may be constrained by the learner’s developmental readiness (Pienemann 1984, 1989, 1998; Williams and Evans 1998; Spada and Lightbown 1999), which may be further mediated by first language (L1) transfer or other Li-based factors (Spada and Lightbown 1999).
As for the relationship between explicit and implicit knowledge, one of the main debates concerns the extent to which explicit knowledge gained through formal instruction can lead to the development of implicit knowledge underlying spontaneous and naturalistic L2 use. Some researchers in the past have argued that there is no relationship between the two forms of knowledge (for example, Krashen 1981, 1985, 1993; Schwartz 1993). However, many SLA researchers now believe that a relationship exists, particularly through: (1) performance of activities that promote the learner’s attention to target forms while processing input (for example, R. Ellis 1982, 1990, 1994a, 1997c, 2003, 2005c; VanPatten 1990; Robinson 1995, 1996, 2001; Doughty and Williams 1998a; White 1998; Williams 2001), or (2) through repeated practice and increased exposure (for example, McLaughlin 1978, 1990; McLeod and McLaughlin 1986; N. Ellis 1994, 1995, 2002a, 2002b; DeKeyser 1998), although N. Ellis cautions (2002b) that rote practice alone does not necessarily facilitate spontaneous TL production, or (3) through making the learning process more efficient by helping learners attend to features in the input that they would not otherwise notice (R. Ellis 1997a, 2003; Doughty and Williams 1998a; Williams and Evans 1998).
Empirical studies on the relationship between explicit and implicit knowledge, however, indicate the existence of a complex relationship between the two types of knowledge suggesting that, while explicit rule learning may be advantageous over implicit rule learning, its advantages depend on a number of linguistic and psycholinguistic variables. For example, some studies have found that explicit instruction may be more effective than implicit instruction when learning involves simple rules (for example, DeKeyser 1995; Robinson 1996; de Graaff 1997). Other studies suggest that the relative benefits of explicit instruction may be more related to factors such as the extent of instruction, the kind of task involved, the amount, nature and timing of planning (see articles in R. Ellis 2005c), the learners’ differences in their cognitive abilities, their stages of L2 learning, frequency effects (see N. Ellis 2002a, 2002b; R. Ellis 2002a, 2002c), and even L2 learning