Understanding Second Language Process

ZhaoHong Han, ZhaoHong Han

• Publisher: Multilingual Matters
• Year: 2007
• ISBN:
  9781788920605
• Topic: Lingue e linguistica
• Subtopic: Psicolinguistica

This book assembles 11 analytical and empirical studies on the process of second language acquisition, probing a wide array of issues, from transfer appropriate processing to L2 default processing strategies, among hearing or deaf learners of a variety of target languages including English, Japanese, Chinese, Korean, French, Spanish, and American Sign Language. Although instruction per se is not the focus of this volume, the chapters are written with instructed learners in mind, and hence offer valuable insights for both second and foreign language researchers and practitioners.

Chapter 1

Revisiting the Role of Consciousness with MOGUL

MICHAEL SHARWOOD SMITH

Thirty years of research has not produced any really hard evidence that making people aware of formal features of the second language (L2) has any significant long-term effect on their grammatical development. However, people still have a persistent feeling that metalinguistic ability in the L2 is more than just a luxury extra or, viewed more pessimistically, more than a distraction and an encumbrance. It is surely a prerequisite for any proper research into such issues that we have a much more fine-grained explanation of the mechanisms involved in metalinguistic ability than has been the case so far. At the least, we need to develop a coherent theoretical model of this ability that we can use to generate interesting research questions about such issues as input enhancement (see Berent & Kelly, this volume) and focus on form (see Han, this volume). You might say that, although there has been no dearth of empirical research, not all that much has happened in this theoretical arena since the 1970s. The MOGUL¹ framework being developed by Sharwood Smith and Truscott aims, among other things, to rekindle the search for more coherent conceptualisations of the problems involved.

MOGUL is a processing model that is devised in such a way as to engage coherently with research across a variety of domains. Following proposals by Ray Jackendoff, it involves a recognition of the existence of a separate, modular language faculty, containing the core phonological and syntactic systems. It also recognises the crucial importance of ‘conceptual structure’, which includes the vital semantic and pragmatic dimensions of language that, in this framework, lie outside this core and allow for the possibility of conscious introspection. In fact, it is in the conceptual domain that metalinguistic ability is anchored, allowing the language user to construct fragmentary or even quite sophisticated metagrammars that co-exist with the inaccessible grammar(s) processed inside the core language modules. Although accessible to conscious awareness, these metalinguistic systems can in principle be recruited in performance skilfully and quite spontaneously.

MOGUL has a way of explaining the relationships between these two types of grammar and the way they may develop and interact, which takes us well beyond the original innovative and widely disputed model proposed by Stephen Krashen and in ways that accord with current research into cognition. It also places metalinguistic ability in a wider context as something that is a regular part of everyday use in both first language (L1) and L2, that is, both in and outside the classroom.

The Early Years

At the close of what might be called the first years of L2 acquisition starting with Corder’s seminal paper on error analysis and ending with Dulay, Burt and Krashen’s Language Two in 1982, three basic ideas had been introduced to the field (Corder, 1967; Dulay et al., 1982):

(1) The learner system;

(2) The ‘developmental imperative’; and

(3) Dual Knowledge Hypothesis.

The first of these was the idea that language learners operate a non-native version of the target language, which one could regard in some sense as systematic and autonomous, in other words, as not merely an ill-assorted collection of correct and incorrect ideas, rules or principles concerning the properties of the L2 in question.

The second basic notion was what one might call the ‘developmental imperative’, namely that the L2 grammar, or a goodly part of the morphosyntax, unfolds in the learner in pre-ordained ways, usually to be interpreted as a pre-ordained sequence or perhaps a set of sequences, such that no intervention by either teacher or learner could influence the course of events, provided that the learner continued to be exposed to the L2. It is as though there is a programmed sequence of steps that needs only exposure to the appropriate input to trigger growth. Aspects of the learner’s system must develop in their own time and in a manner much constrained by in-built principles (not provided by the environment) – it is the role of SLA research to determine the nature and scope of these principles. Although the idea of fixed sequences is not a Chomskyan notion nor is it necessarily implied by the existence of principles of universal grammar, the basic idea fits in neatly with the general assertion, familiar from generative linguistics, that because mother tongue (L1) grammatical development unfolds without the need for correction or introspection, it must therefore in some way be helped along by innately given principles. The claim put forward by various L2 researchers was, of course, that L2 development is driven by essentially the same principles, although their implementation might in practice encounter more obstacles on the way.

The third and perhaps most contentious notion was that learning language in a conscious, analytic manner resulted in a kind of knowledge that was quite distinct from that which underlies spontaneous language use. Knowing the language is not the same as knowing about the language. SLA theory must account for the characteristic discrepancy between the two.

The nub of the problem

The most contentious part of the dual knowledge idea was Krashen’s assertion that there was no ‘interface’ between consciously and subconsciously gained grammatical knowledge (Krashen, 1976, 1978, 1985; Sharwood Smith, 1981). Conscious rules could not be converted into subconscious ones, no matter how hard one practised. If this were so, then the developmental sequences could easily be subverted and, all other things being equal, particular training programmes could dictate the actual sequences in which learners gained control over given areas of the L2 grammar. The research carried out during this early period suggested that conscious learning could not interfere with the pre-ordained sequences of development and only made the learners more knowledgeable about the nature of the ‘errors’ they committed. The best thing was to focus on the meaning and not the form of the message (Dulay et al., 1982).

Looking back on these early days from a 21st century perspective, one can see that these three basic ideas have stood the test of time. Even the conscious/subconscious distinction as presented in Krashen’s various publications is still a live issue, and it took until 1998 for a collection of studies to come out that signalled a concerted attempt to find support for drawing learners’ attention to formal aspects of the L2, and this included explicit reference to (or ‘focus on’) form as opposed to meaning (Doughty & Williams, 1998a; Long & Robinson, 1998). Moreover, Doughty herself, in a state-of-the-art paper in 2003, expresses great reserve about the role of consciousness, that is, ‘metalinguistic’ knowledge in grammatical development (Doughty, 2003:298). Although Schmidt claims that no learning can take place without noticing, the precise nature of noticing, whether it involves some degree of conscious awareness – without necessarily implicating analytic understanding (Schmidt, 1990) – is constantly under scrutiny and is no way decided (Carroll, 2001; Truscott, 1998).

After so many years of empirical research, one senses in the SLA literature a growing trend to refine the conceptual basis for making claims about how an L2 develops over time and, in particular, how the basic ideas outlined above accord with contemporary research in sister disciplines. In the discussion that follows, the idea surrounding ‘metalinguistic’ ability and its role in L2 development will be revisited, this time using the MOGUL framework, the aim behind which is precisely to achieve the required increase in theoretical precision and cross-disciplinary validity.

Dual knowledge: The essentials

Krashen’s conscious/subconscious learning distinction can be explicated as follows:

(1) Conscious learning produces ‘technical’ (‘meta’) knowledge about the target language.

(2) Using this knowledge requires the operation of a conscious mental editor called the ‘Monitor’.

(3) The Monitor is a grammatical ‘first aid kit’. It is

(a) linguistically unsophisticated (for most learners);

(b) available only for non-spontaneous use; and

(c) not always used anyway, even in optimal circumstances.

What are the consequences for language learners (and teachers)? Put in its most extreme form, the advice to learners is to throw away the grammar book and just try to listen, read, and understand. Learning words and phrases, in the sense that this is focusing on meaning and not on the systematic aspects of L2, is the only guaranteed way in which a conscious approach to L2 learning will bring benefits. In short, the role of conscious learning is given at most a very peripheral role.

In the 1970s, the evidence for the no-interface hypothesis was not solid and hence open to considerable scepticism. However, despite the failure to establish a strong empirically backed foundation for conscious learning in defiance of the no-interface hypothesis, people continue to have a persistent feeling that metalinguistic ability in the L2 is more than just a luxury extra that is of no real relevance to L2 development. This may be justified, but where is the hard evidence for this feeling? Thirty-five years of research has not produced any substantial proof that making people aware of formal features of the L2, whether by means of correction or explanation or both, has any long-term effect, at least where basic morphosyntax and phonology are concerned.

Note that it is important to mention what aspects of the language we are talking about, because when talking about ‘the’ language or ‘the’ L2, people sometimes ignore substantial aspects of language that (1) have not been as well researched as syntax and phonology and (2) may by their very nature be amenable to manipulation by external agents, either teachers or the learner ‘outside’ the inaccessible parts of language, that is, by the conscious learner (see relevant discussion in Sharwood Schwartz, 1999; Smith, 1994). The role of ‘conscious’ learning outside core syntax and phonology has never been explicitly denied but the issue has remained vague and relatively unexplored. It also depends on what you mean precisely by ‘lexical’ and ‘pragmatic’ aspects of L2. You certainly need a proper linguistic theory to determine this: the empirical data will not tell you and traditional grammatical categories are notoriously vague and unreliable. Is the English definite article ‘the’ a functional/grammatical phenomenon or a lexical one, for instance? The category ‘article’, or ‘determiner’ or Det are functional terms but the particular items in English that actually fulfil this morphosyntactic function can easily be called lexical, and the selection of which article should go in which context may well be dictated by semantic or pragmatic principles. In sum, the question of conscious awareness and what aspects of language it has absolutely no access to remains a problematic and complex issue that requires both linguistics and other disciplines to be brought to bear on it.

Alternatives to dual knowledge

There are various theoretical perspectives that allow for different modes of knowledge but deny total separation between them (DeKeyser, 2003; Hulstijn, 2002). Development in one area may contribute to development in the other, as knowledge is ‘automatised’ or reformatted for real-time processing. There are also theoretical perspectives that would, in addition, account for language development without resorting to notions like ‘knowledge’ or ‘internal representations’ of language at all (N. Ellis, 2003). Realistically, these approaches still have to make their case and provide enough evidence to discredit their rival and more established approaches concerning the acquisition of the core language system, that is, syntax and phonology. In the meantime, we have to assume they contain some interesting insights but cannot yet account for how the totality of an L2 system is acquired. In sum, the jury is still out.

Limits of Krashen’s explanations

Whereas Krashen could have referred us to linguistic theory as far as subconscious knowledge of language was concerned, he could not say much about the mechanisms of conscious learning given the relative dearth of research then in this still thorny area of cognitive science. He also appeared to assume a sharp, black-and-white distinction between conscious and non-conscious modes of processing.

Monitor Theory was an interesting but sketchy claim about how we gain and use conscious knowledge and it was, so its critics asserted, substantiated only by a bare minimum of hard and hotly contested evidence. Also, as regards development, developmental sequences were pretty much magic; no overall explanation was forthcoming at the time. We had to wait for researchers working on German migrant worker data for the first attempt to really explain sequences (e.g. Clahsen, 1984; Meisel et al., 1981; Pienemann, 1984). On the credit side, Krashen’s claims were at least interesting, relevant and made a wide impact at the time, and so they have posed useful challenges for researchers that are still unanswered at the time of writing.

Other approaches to metalinguistic ability

Various researchers expressed critical views concerning Krashen’s claims. In 1981, for example, Sharwood Smith responded by saying that what he called grammatical ‘consciousness-raising’ (C-R) as presented by Krashen was too simplistic and that research into the effectiveness or non-effectiveness should proceed with a more fine-grained conceptualization that defined C-R not as a rigid dichotomy but as a range of options along two dimensions: one which had to do with degrees of explicitness ranging from the very subtle, indirect cues that might attract the learner’s attention to the very explicit, where grammatical forms were pointed out and talked about, and another dimension, namely elaboration, which described how much time and space was devoted to the C-R activity by teacher or textbook. Only when research had investigated all the various possibilities allowed by this scheme could we say definitely that drawing the learner’s attention to grammar rather than meaning was always ineffective. Sharwood Smith also speculated that learners might be able, metalinguistically, to generate their own input by consciously constructing L2 utterances, use them to communicate, and then acquire the grammar subconsciously according to principles as outlined by Krashen alongside other grammatical forms that were embedded within utterances addressed to them by other speakers (or writers), that is, L2 input in the standard sense of the term.

Later, in 1993, Sharwood Smith proposed that C-R should be replaced by input enhancement in that it focused on what we know, namely that we have manipulated the visible or audible input in some way (any of the ways described along the two dimensions of explicitness and elaboration), and did not presume to describe in advance the actual psychological effect on the learner: what is salient in the input and salient for the observer will ultimately depend on internal mechanisms in the learner.

The 1990s

As evidenced by the Doughty and Williams (1998a) book of readings, research into these issues finally got under way in earnest in the second half of the 1990s. The terminology adopted by many for this was based on Long’s formulation of the difference between focus on form(s) and focus on form, the former being the traditional explicit and elaborate focus on grammar and the latter being a strategic and sporadic focus, presumably relatively unelaborated and with varying degrees of explicitness, firmly embedded within a communicative, that is, meaning-oriented, approach to language instruction (Long & Robinson, 1998).

The limitations of explicit focus on form

Despite the promise of breakthroughs, research into focus on form has been plagued with methodological obstacles: for example, too many variables, absence of late post-tests, and so on (see the critique in Trenkic & Sharwood Smith, 2002). It is possible to say in general that some form of explicit instruction has been shown to facilitate learning with regard to given areas of the L2. In fact, the research literature is full of encouraging signals and ‘promissory notes’ that would appear to go beyond the objective experimental findings, but without a better theoretical backing, researchers risk generalising from short-term effects to long-term success. Varying degrees of doubt have certainly been voiced. Truscott, for example, concludes in 1997 that ‘research on form-focussed instruction has produced essentially no evidence that it is helpful and has produced considerable evidence that it is ineffective (though the latter is not entirely conclusive)’ (Truscott, 1996, p. 121) and Doughty (2003) concludes her review of instructed SLA with a proposal advocating an indirect, implicit approach to focusing on form, saying that instruction should help to ‘organise the learner’s processing space’ (2003: 298) so that the (adult) learner can pick up on the cues the same way young learners can.

MOGUL: An Overview

Conceptual structure and cognition in general

The MOGUL framework being developed by Sharwood Smith and Truscott aims, among other things, to fill this gap. MOGUL is a real-time processing model that is devised in such a way as to engage coherently with research across a variety of domains.

Following proposals by Ray Jackendoff, it involves a recognition of the existence of a separate, modular language faculty, containing the core phonological and syntactic systems, each of them also being modular in character. It also recognises the crucial importance of ‘conceptual structure’, which includes the vital semantic and pragmatic dimensions of language, which, in this framework, lie outside this core and allow for the possibility of conscious introspection. Conceptual structure, which provides the support system for our complex mental life (e.g. Jackendoff, 1987; Pinker, 1997), is also linked with representations that have their basis in the senses: sounds, tactile and olfactory sensations, and visual images. All of this is best illustrated by considering dreams where not only thoughts are invoked, but also a multitude of sensations that have no basis in what is happening outside us. In other words, sounds and images have an independent existence in our minds and the virtual sensory experiences we have must be supported by something that triggers them from ‘inside’ rather than from outside in the external environment. The necessary perceptual (visual, auditory, and so on) representations that fulfil this role will be crucial when the issue of metalinguistic knowledge is considered.

Constraints imposed upon conscious introspection

First, the basic ways in which we process and acquire language according to MOGUL will be briefly considered (for a fuller account, see Truscott & Sharwood Smith, 2004a...

Table of contents

• Cover
• Title page
• Copyright
• Contents
• The Contributors
• Preface
• Chapter 1: Revisiting the Role of Consciousness with MOGUL
• Chapter 2: Multi-Competence: Black Hole or Wormhole for Second Language Acquisition Research?
• Chapter 3: Transfer Appropriate Processing as a Model for Classroom Second Language Acquisition
• Chapter 4: On the Role of Meaning in Focus on Form
• Chapter 5: The Efficacy of Visual Input Enhancement in Teaching Deaf Learners of L2 English
• Chapter 6: Learner Spontaneous Attention in L2 Input Processing: An Exploratory Study
• Chapter 7: Working Memory and L2 Processing of Redundant Grammatical Forms
• Chapter 8: L2 Learners’ Interpretation of Operator-Variable Binding in VP Ellipsis
• Chapter 9: Metasyntactic Ability in L2: An Investigation of Task Demand
• Chapter 10: Prosody Acquisition by Japanese Learners
• Chapter 11: Recognition and Production of Formulas in L2 Pragmatics
• References
• Index