An Evaluation Of The Study Of Sla

The field of Second Language Acquisition (SLA) is a relatively young one and is therefore have said to have 'teething problems' (Doughty and Long, 2003: 4). The two main issues of the study of SLA are its youth and its interdisciplinary nature. These issues, which are not necessarily weaknesses, as I will argue later, have led to a proliferation of theories in the field, with authors counting as many as 60 (Long, 2007: 4). In an attempt to evaluate SLA as an academic enterprise the purpose of this paper is twofold. First, the main topics of research will be discussed with reference to the history of SLA. Second, the maturation of theory construction in SLA will be examined.

The first theory to be extensively applied to SLA has its origins, not in SLA, but in psychology. In behaviourism, language learning was viewed as forming new habits. Although widely applied to language pedagogy, the research supporting the theory was very scarce. The belief that new L2 habits had to be formed led to Contrastive Analysis, in which the L1 was compared in detail to the L2 in order to predict learners' difficulties in L2 learning (Mitchell and Myles, 2004: 32).

The 1950s and 1960s saw a cognitive revolution which marked a turning point in linguistics (Hulstijn, 2007: 192). With the emergence of psycholinguistics, based on the Chomskyan view of an innate, modular, language faculty, research focused on first language acquisition, especially with regard to the input and the order of acquisition. Similar research into the order of acquisition of morphemes was soon conducted in the study of SLA. Errors were no longer viewed as mistakes, but as significant insights into the 'interlanguage' of the learner (Corder, 1967).

The significance of errors gave rise to Error Analysis. Dulay and Burt (1973) and Bailey et al. (1972) used Error Analysis to demonstrate that the majority of errors made by learners could not be explained by the role of the L1 (Mitchell and Myles: 38). This constituted a considerable attack on Contrastive Analysis. Ellis points out that CA research (e.g. Tran-Chi-Cahu in 1974) claimed as much as 51% of learners' errors were due to L1 interference, whereas Dulay and Burt asserted it was only 3% (1985: 29).

In the 1980s researchers of interlanguage also began to focus on its processes: simplification, overgeneralization, restructuring, and U-shaped behaviour (Ortega, 2009: 116-8). Proponents of Universal Grammar set out to show systematicity in learner language development in the acquisition order of morphemes and syntactic structures. Although some linguists view the findings on developmental sequencing as robust, others have "expressed doubts on the value of past interlanguage findings because they are based on the metaphor of 'a developmental ladder'" (Larsen-Freeman, 2006: 594 in Ortega, 2009: 141).

Simultaneous to the cognitive turning point in linguistics, psychologist and neurobiologist Lenneberg conducted research into lateralization in children and its effects on language development. This resulted in the popularization of the critical period hypothesis, which led to a number of studies into first and second language acquisition with regard to the 'end' of this critical period (Obler and Gjerlow, 1999: 66). The fact that SL learning is often typified by incomplete success (Ortega, 2009: 18) became of theoretical interest. "This 'stopping short' has been referred to as fossilization (Selinker, 1972) or incompleteness (Schachter, 1990). It is one of the noticeable characteristics of SLA." (Towell and Hawkins, 1994: 2).

Other approaches also gained credibility in the 1970s and 1980s. Sociolinguistic views on SLA looked at social context of language learning. In Schumann's acculturation model, for example, the degree in which a learner feels part of the community of the target language was seen as an important factor, as well as matters of integration and attitude (Larsen-Freeman and Long, 1991: 253). The primary research focus was on variation and incompleteness (Towell and Hawkins: 42).

In summary, the main topics of research in SLA from roughly the 1950s up until the 1990s has included the role of L1, analysis of errors, the role of input and frequency, interlanguage processes, variability and systematicity in interlanguage, (dis)similarities to first language acquisition, age effects, incompleteness and sociopsychological factors. All these topics remain important today. The progress in the field in the past twenty years will now be looked at by discussing the main strands of theories and the research conducted therein.

At this point it may be useful to make a distinction between symbolic and subsymbolic theories of SLA. In symbolic theories of language acquisition, such as nativist accounts, "knowledge is represented by means of symbols and operators or rules that specify the relationships between symbols" (Hulstijn, 2002: 196). In subsymbolic theories, such as connectionism, the representation of knowledge is distributed "as a pattern of activation in a neural network containing hidden units" (Hulstijn, 2002: 196). In other words, the latter type of theory assumes the neural architecture of the brain allows it to carry out multiple complex operations (Randall, 2007: 18). The focus is therefore on the mechanisms of learning.

From a generative, or symbolic, perspective, "linguistic competence is an abstraction" (White, 2007: 43) and research conducted will attempt to tap into unconscious knowledge, for example through elicited production. The most commonly used technique is the grammaticality judgement task. White and Juffs (1998) used these two techniques to investigate whether L2 learners of English observe island constraints in order to demonstrate that learners' grammars adhere to UG principles (White: 47-50).

In a cognitive connectionist approach, evidence can come from very diverse data, such as "educational experimental designs, psycholinguistic studies of sentence processing, analyses of learner production, and brain imaging and neuroscience" (Ellis, 2007: 85). In their study, Goldschneider and DeKeyser (2001) analysed the oral production data from twelve different morpheme order studies as well as input frequency from a spoken English corpus in order to reframe the past findings and argue for an explanation of associative learning based on usage-based acquisition (Ellis: 85-7).

Some approaches attempt to explain learning processing by placing value on the roles of input, output, and/or interaction. For example, based on her observations on the output of children in immersion programmes in Canada, Swain developed the Output hypothesis, which states that the actual act of speaking forces learners to try out how the grammar of the L2 works and will also invite feedback from interlocutory partners (Mitchell and Myles: 21).

VanPatten's theory of Input Processing is concerned with how adult L2 learners process the input. Aspects considered important in this process are the making and mapping of form-meaning connections and the parsing of input sentences (VanPatten, 2007: 125). Research designs used to examine these aspects are sentence interpretations tasks and eye tracking (126). In a study by VanPatten and Houston (1998), the researchers manipulated a collection of sentences to find out how subjects would interpret embedded OVS structures (129-30).

Gass and Mackey's Interaction Approach is based on the assumption that all three components (input, output and interaction) bring about language learning (2007: 181). Research in this vain would focus on the role of feedback, negotiation for meaning, language-related episodes and attention in different tasks (181-189). A study conducted by Mackey et al. (2000) focussed on how L2 learners perceive feedback during interaction by subjecting learners to communicative tasks, followed immediately by stimulated recall methodology to collect introspective data (Gass and Mackey, 2007: 190-192).

Some theories look at the role of cues in language acquisition. In Carroll's nativist Autonomous Induction Theory, cues are important in explaining how structural distinctions are made in the learner's structural representations (2007: 158). In MacWhinney's emergentist Unified Model, cue strength is studied by means of direct-object identification tasks, relative clause processing tasks, and pronominal assignment tasks (2008: 354). Cue validity is measured by text counts based on relative availability and reliability of the cues.

The purpose of this paper is not to assess every individual theory in SLA, but rather to evaluate the study of SLA as a whole. It is for this reason that the above discussion on main focal points of research in the past two decades is by no means a comprehensive one. Rather, it aims to provide a general overview of what has been studied, why, and how. It can be concluded that the past twenty years have seen a proliferation of theories with diverse research focal points. The question that remains is the value of theory in SLA.

To truly do this question justice, it is of interest to consider why linguists and academics from other disciplines study SLA at all. Mitchell and Myles argue for two basic reasons. The first reason is the intrinsic value of a better understanding of the nature of language, the human mind, human learning, and communication. The second reason SLA should be studied is its applicability to L2 learning and teaching (6). Relating pedagogy to academic research is an issue beyond the scope of this paper, but I will follow the view that the 'real world of application' is at serious risk if fundamental questions are not continued to be addressed (Hulstijn, 2007: 196), as research findings "often affect the life chances of 'at risk' populations" (Larsen-Freeman and Long: 290).

Having defined the purpose of the study of SLA, I am now able to question the role of theory construction in the field. Most authors agree the role of theory is essential in improving our understanding of SLA because theory motivates research (Larsen-Freeman and Long: 222; Mitchell and Myles: 7). Larsen-Freeman and Long point out that, unfortunately, theory is often accorded low priority which leads to less useful research (222).

Long is concerned that the field of SLA "has become increasingly fragmented during the past 15 years, and is characterized by a multiplicity of theories", which he sees as a result of the various disciplinary backgrounds (2007: 3). He further argues that the proliferation of oppositional theories obstructs progress (14). This is in opposition to Jordan's view, who writes that "as many theories as possible should be encouraged" (116). Although Long agrees with Jordan that SLA is "too broad a field to be handled by any currently available single theory", he is worried that theory proliferation will ultimately render SLA unaccountable (27).

I certainly think a monolithic and powerful theory of SLA will not advance the field. It is of essence that research is conducted within multiple theoretical frameworks to provide robust findings. It is also of great importance that SLA researchers can no longer be naive about scientific research and perhaps there is a need for 'gatekeepers' in SLA literature to achieve greater accountability (Long: 166). In order to unify the field, theories should focus on the fundamental issues in SLA (Hulstijn, 2007: 200). Any SLA theory should only gain credibility when evaluated according to certain criteria, for example Jordan's criteria, which include the need for theories to be coherent and consistent, to have empirical content, to be fruitful, broad and simple (116-7).

There is disagreement as to what the fundamental issues in SLA are. In the Appendix I have made a table with an overview of the core issues according to Hulstijn (2007), VanPatten and Williams (2007), Long (2007), Mitchell and Myles (2004) and Towell and Hawkins (1994). There is no issue which is unanimously agreed upon. This lack of agreement on what should be investigated typifies the disintegration of the field. For example, the terminology of the authors is different. Where Mitchel and Myles use general descri ptions, for example, 'the role of the input', Long is very specific in describing, for example, 'autonomous interlanguage syntax'. In my opinion, the field would benefit from slightly more broadly defined issues, without a theoretical bias in describing them.

In the appendix, I have shaded those phenomena grey I believe need to be addressed in theories of SLA. Incidental learning is very much debatable and is therefore left out. Similarly, variability across subsystems is too linguistic a view on language acquisition. There are no true boundaries between, for example, morphology, syntax and semantics. The role of instruction is left out deliberately as well, as the findings will need to be thoroughly established before thinking of pedagogical implications. The value of instruction as such is irrelevant in this sense, because instruction is a fact for many language learners. What needs to be investigated at a later stage is how instruction may be improved, based on the research conducted.

Should a single theory be able to account for all observable phenomena? I think not. Researchers must take caution in labelling and defining the phenomena they wish to explain in order to provide plausible explanations and allow for across-theory evaluation.

Finally, I believe the youth and interdisciplinary nature of SLA are not weaknesses, but strengths. That is to say, they have certainly led to problems, such as loose theoretical definitions or lack of empirical content, but simultaneously have provided the field with interdisciplinary knowledge it would not have been able to incorporate had it been older, established and autonomous. I would also like to point out that the mechanisms of learning have become important and that these deserve equal attention to, or possibly more than, that already paid to the representation of knowledge alone. There is still a great deal to be discovered about implicit and explicit learning. Also, the advances made in the field of cognitive science, a field which, like SLA, is young and interdisciplinary (Doughty and Long: 869), will help SLA researchers further their understanding of the human mind. References

Carroll, S. (2007). Autonomous Induction Theory. In VanPatten, B. and Williams, J. (Eds.), Theories in SLA: An Introduction,155-173. New York and London: Routledge. Corder, S.P. (1967) 'The Significance of Learner's Errors'. International Review of Applied Linguistics 5: 161-170. Doughty, C., & Long, M. (Eds.) (2003). The Handbook of Second Language Acquisition. London: Blackwell. Ellis, N. (2007). "The Associative-Cognitive CREED" . In VanPatten, B. and Williams, J. (Eds.), Theories in SLA: An Introduction, 77-95. New York and London: Routledge. Ellis, R. (1985). Understanding Second Language Acquisition. Oxford: Oxford University Press. Gass, S. and Mackey, A. (2007). 'Input, Interaction, and Output in Second Language Acquisition'. In B. VanPatten & J. Williams (Eds.), Theories in Second Language Acquisition: An Introduction, 175-199. New York: Routledge. Hulstijn, J. (2002). 'Towards a unified account of the representation, processing and acquisition of second language knowledge'. Second Language Research 18: 193-223. Hulstijn, J. (2007). 'Fundamental issues in the study of second language acquisition'. EUROSLA Yearbook 7: 191-203. Jordan, G. (2004). Theory Construction in Second Language Acquisition. Amsterdam: Benjamins. Larsen-Freeman, D. & Long, M. H. (1991). An Introduction to Second Language Acquisition Research. London: Longman. Lightbown, P.M. & Spada, N. (2006). How languages are learned, 3rd edition. New York: Oxford University Press. Long, M. (2007) Problems in SLA. Mahwah, New Jersey: Lawrence Erlbaum Associates, Inc. MacWhinney, B. (2008). 'A Unified Model'. In P. Robinson & N.C. Ellis (Eds.), The Handbook of Cognitive Linguistics and Second Language Acquisition, 341- 371. New York: Routledge. Mitchell, R. & Myles, F. (2004). Second language learning theories. Second, revised edition. London: Arnold. Obler, L. and Gjerlow, K. (1999). Language and the Brain. Cambridge: Cambridge University Press. Ortega, L. (2007) 'Second Language Learning Explained? SLA Across Nine Contemporary Theories'. In B. VanPatten & J. Williams (Eds.), Theories in Second Language Acquisition: An Introduction, 225-250. New York: Routledge. Ortega, L. (2009). Understanding Second Language Acquisition. London: Hodder Education. Randall, M. (2007). Memory, Psychology and Second Language Learning. Amsterdam, Philadelphia: John Benjamins Selinker, L. (1991). Reframing Interlanguage. London and New York: Longman. Towell, R. & Hawkins, R. (1994). Approaches to second language acquisition. Clevedon, UK: Multilingual Matters. VanPatten, B. (2007). 'Input Processing in Adult Second Language Acquisition'. In B.VanPatten & J. Williams (Eds.), Theories in Second Language Acquisition: An Introduction, 115-135. New York: Routledge. VanPatten, B. and Williams, J. (2007). Theories in Second Language Acquisition: An Introduction. New York: Routledge. White, L. (2007). 'Linguistic Theory, Universal Grammar, and Second Language Acquisition'. In B. VanPatten & J. Williams (Eds.), Theories in Second Language Acquisition: An Introduction, 37-55. New York: Routledge.