Strategies for Language Learning and for Language Use: Revising the Theoretical Framework

ERNESTO MACARO Department of Educational Studies University of Oxford Oxford OX2 6PY United Kingdom Email: ernesto.macaro@edstud.ox.ac.uk Since the late 1970s, there has been widespread research interest in the strategies that learners use in learning and using second languages. This interest has generated a parallel research effort in language learner strategy instruction. The body of work to date suggests a possible relationship between strategy use and second language learning success. It also provides some evidence that learners can be helped to use strategies more effectively. Several criticisms, however, have been made of this eld of research, particularly pertaining to a lack of theoretical rigour. This article reviews the problems related to strategy research and proposes a revised theoretical framework in which strategies are differentiated from skills, processes, and styles. Rather than offering an all-encompassing denition of a strategy, the article proposes a series of features essential to describing a strategy. The framework aims to enhance current theory.

THE LAST 25 YEARS OR SO HAVE WITNESSED considerable growth in research activity in learner strategies. This research has attempted to explore the strategies that learners of a second language (L2) deploy either when learning a language or when using a language or both (Cohen, 1998). A great deal of this research effort owes much to four often-cited books (Naiman, Frohlich, Stern, & Todesco, 1978/1996; OMalley & Chamot, 1990; Oxford, 1990; Wenden & Rubin, 1987) which, taken together, attempted to provide the theoretical underpinnings for future research and demonstrated the direct applicability and appeal of learner strategies to classroom practitioners. This article is divided into ve sections. In the rst section, I provide a summary of the claims made by researchers operating in the eld of learner strategies. In the second section, I bring together a range of criticisms of learner strategy research. In the third section, I explore previous theoretical frameworks in order to ascertain the

extent to which these criticisms are justied, focusing particularly on construct validity. In the fourth section, I offer a theoretical framework that seeks to overcome the problems identied in previous sections. In the fth section, the discussion, I summarise the advances the theoretical framework seeks to promote while proposing ways of testing the framework empirically. CLAIMS MADE IN LEARNER STRATEGY RESEARCH The theorising and the practical applications offered by the four previously mentioned books, together with the empirical studies that have followed them, have yielded a body of evidence leading scholars to make the following claims:1 1. Strategy use appears to correlate with various aspects of language learning success. Some studies show correlations between generally high strategy use and learning success (see Oxford & Burry-Stock, 1995, for a review) or between generally high strategy use and motivation (Nunan, 1997; Oxford & Nyikos, 1989). Some studies show a link between success and a preference for

The Modern Language Journal, 90, iii, (2006) 0026-7902/06/320337 $1.50/0 C 2006 The Modern Language Journal

Ernesto Macaro certain kinds of strategies that mark good language learners (Naiman et al., 1978/1996). For example, Beaton, Gruneberg, and Ellis (1995) found that vocabulary was very effectively retained by the use of the keyword strategy. Gu and Johnson (1996) found that visual repetition (writing words down repeatedly) negatively correlated with vocabulary size, whereas selective attention and selfinitiation correlated positively. N. J. Anderson (1991), Block (1986), Carrell (1989), and Devine (1984) all found that successful readers deployed a group of strategies that integrated meaning in preference to surface text-based strategies. Some studies have shown a link between success and combinations of strategies; these in turn being allied to metacognition and to exibility of deployment (Chamot & El-Dinary, 1999; Graham, 1997; Macaro, 2001; OMalley, Chamot, & Kupper, 1989; Sanaoui, 1995; Vandergrift, 1998; Vann & Abraham, 1990). A later section of this article will return to this apparent lack of consensus among researchers in the eld as to whether it is the range and frequency of strategy use, the nature of strategies, or the combinations of strategies that is the key to successful language learning. 2. There are group differences and individual differences in learner strategy use. Females seem to use more strategies than males (Ehrman & Oxford, 1989; Macaro, 2000; Oxford & Niykos, 1989; Sheorey, 1999), or they use strategies differently (Bacon, 1992; B ugel & Buunk, 1996; Gu, 2002). Cultural groups may vary in their strategy use (Kim, 1999; Koda, 1990; Levine, Reves, & Leaver, 1996; LoCastro, 1994). Experienced L2 learners may use sets of strategies different from those of inexperienced L2 learners (De Larios, Murphy, & Manchon, 1999; van Hell & Mahn, 1997). Individuals may restrict themselves to an ineffective subset of strategies (Kember & Gow, 1994; Porte, 1997) or may be unable to deploy a number of strategies (Block, 1986; Knight, 1994; Lee & Schallert, 1997) or may use strategies inappropriately without knowing that they are doing so (Christianson, 1997; Porte, 1995). 3. The methodology for eliciting learner strategy use, although imperfect, is at an acceptable level of validity and reliability. Questionnaires and inventories provide the broad picture; verbal reports (think-aloud techniques and taskbased retrospectives) effectively yield insights into skill-specic or task-specic strategy use. With regard to verbal reports, these claims of effectiveness are made both by L2 strategy researchers (N. J. Anderson, 1991; N. J. Anderson & Vandergrift, 1996; Cohen & Hosenfeld, 1981; Oxford & Burry-Stock, 1995; Phakiti, 2003; Sarig, 1987) and

321 by researchers operating in other elds related to cognitive processing (Bialystok, 1981; Bowles & Leow, 2005; Ericsson & Simon, 1987; Leow & Morgan-Short, 2004; Nayak, Hansen, Krueger, & McLaughlin, 1990; Schraw & Moshman, 1995). 4. Despite some setbacks (OMalley, 1987; Wenden, 1987b), and some reservations (D ornyei, 1995; Lam & Wong, 2000), learner strategy instruction (or training) appears to be effective in promoting successful learning if it is carried out over lengthy periods of time and if it includes a focus on metacognition. For example, a number of studies have described how learners were trained to be more effective writers (Aziz, 1995; Conti, 2004; Macaro, 2001); how learners instructed in communication strategies improved their oral interaction (Cohen, Weaver, & Li, 1995; Holunga, 1994; Nakatani, 2005); and how interventions appeared successful in enhancing vocabulary acquisition (Avila & Sadoski, 1996; Burgos-Kohler, 1991; Cohen & Aphek, 1980; Lawson & Hogben, 1998), in enhancing listening skills (McGruddy, 1999; Ozeki, 2000; Seo, 2000; Thompson & Rubin, 1996), and in enhancing reading skills (Carrell, Pharis, & Liberto, 1989; Fraser, 1999; Raymond, 1993). Other strategy instruction studies have claimed to improve general approaches and attitudes to language learning (Flaitz & Feyten, 1996; Kohler, 2002; Nunan, 1997; Sengupta, 2000; Victori & Lockhart, 1995). CRITICISMS OF LEARNER STRATEGY RESEARCH Despite the apparent enthusiasm in the learner strategy domain and the impression perhaps given thus far of a coherent research agenda, a number of criticisms have been leveled at learner strategy research. Some of these criticisms concern the methodology used to elicit, measure, and classify strategies; some target the methodology used to carry out intervention studies; some focus on assumptions about the impact of strategy use; and some examine the lack of theoretical rigour of learner strategy research generally. With regard to qualitative methods for eliciting learner strategies, Seliger (1983) doubted whether the verbalizations of learners represent some form of internal reality (p. 180). In terms of strategy measurement, LoCastro (1994, 1995) argued that large and general learner strategy inventories (perhaps the best known is the Strategy Inventory for Language Learning [SILL], Oxford, 1990) are not transferable across sociocultural domains and that their results and conclusions

322 might therefore be invalid. Furthermore, D ornyei (2005) has attempted to demonstrate that the SILL is psychometrically awed because the scales in the SILL cannot be considered cumulative. Phakiti (2003) argued that frequency counts and simple analysis of variance (on which a great deal of reliance has been placed) do not capture the interaction of several variables when comparing learner strategies with learning outcomes. With regard to the categorization of strategies, Skehan (1991) concluded that, although there is some consensus on the classication of learner strategies being elicited, there is a need to go beyond convenient classications (p. 287), even though these might be useful for strategy instruction, and make links between classication schemes and underlying theory. In relation to interventions into the strategic behaviour of learners, Hassan et al. (2005), in a systematic review of learner strategy instruction, claimed that there has been little standardisation of either the intervention packages or the outcomes that have been measured. This lack of standardisation presented major difculties to Hassan and her associates in their attempts to synthesize studies and to indicate the weight of evidence about their effectiveness. Kellerman (1991) dismissed learner strategy instruction as irrelevant on the basis that learners have already developed strategic competence in their rst language (L1) and can therefore simply transfer it to their L2. Some authors have critiqued the strong claims made about the link between learner strategy use and language learning success. For example, Gillettes (1994) study of three effective and three ineffective learners led her to question the belief that positive learning strategies (p. 211) constitute a full explanation for L2 achievement and that this belief ignores motivation and personal histories. Rees-Miller (1993) referred to a number of unsuccessful interventions and argued that there was, as yet, no demonstrated causal relationship between strategy awareness and L2 learning success, that few strategies were transferable beyond a specic task, and that not all strategy users appeared to be, or to become, good learners of the L2. However, her strongest attack on strategy research, and the one that most concerns this article, is the lack of clarity in the denition of what a strategy actually is. She claimed that:
Even the cognitive learning strategies, such as seeking meaning, using deduction, inferencing, or monitoring, are dened so broadly that it is questionable whether they can be specied in terms of observable, specic, universal behaviours that could be taught to or assessed in students. (p. 681)

The Modern Language Journal 90 (2006) Other researchers had already alluded to ReesMillers concerns about strategy denition. Bialystok (1983), more than 2 decades ago, argued that there was little consensus in the literature concerning either the denition or the identication of language learning strategies (p. 100). Stevick (1990), in an examination of previous definitions, alluded to two major problems. First, there was the problem of what he classied as the Size-Abstractness Dilemma (p. 144), which demonstrates that some strategies refer to phenomena that are larger than others and that some strategies refer to phenomena that are more abstract than others, variations that lead to imprecision and confusion. Second, for Stevick, there was the Outside-Inside Problem, which demonstrates that no clear relationship exists between external acts and the mental constructs to which they are attributed (p. 144). Although concluding that learner strategy research was at an embryonic stage and that we were dealing with a clear example of a researchthen-theory perspective (Skehan, 1989, p. 98), Skehan still remained optimistic as to its usefulness. His conclusions 14 years later were more pessimistic; D ornyei and Skehan (2003), reviewing denitions of strategies, argued that a strategy cannot be either cognitive or emotional or behavioural. They asked whether a strategy was a neurological process, a cognitive operation, or a behavioural act involving motor skills. They questioned, moreover, whether a strategy could contribute to both knowledge and language skills and posited that there was no theoretical explanation for how strategies might be related to skills. They concluded that a theoretical basis for the concept of learner strategies was still sadly lacking and that to provide a scientically rigorous denition researchers would have to provide a coherent neurological and biological account of behaviours, something that D ornyei and Skehan considered an enormous undertaking. In his most recent attack on learner strategies, D ornyei (2005) catalogued the inability of researchers to explain the difference between engaging in an ordinary learning activity and a strategic learning activity (p. 164), a problem that has led him to question the very existence of learner strategies.

PREVIOUS ATTEMPTS AT A THEORETICAL FRAMEWORK OF LEARNER STRATEGIES We will now examine the extent to which researchers in the eld have attempted to address the issue of theoretical rigour in general and strategy denition in particular.

Ernesto Macaro There have been repeated attempts to dene strategies by situating them in psychology, within theories of cognition. Wenden (1987a) viewed strategy research as part of the general area of research on mental processes and structures that constitute the eld of cognitive science (p. 6). From this perspective, one would expect strategies to be located in the brain. However, Wenden dened strategies in terms of language learning behaviours that learners engage in, the knowledge they have of their behaviours, and the knowledge they have of themselves as learners and the language they are learning. We should note the lack of clarication as to whether we are still dealing here with mental behaviour, or whether overt action is also envisaged. OMalley and Chamot (1990) also situated learner strategies within cognitive theory, laying their framework over J. R. Andersons (1980/ 2000) ACT model and Stages of Skill Acquisition model. Again, the expectation would be that strategies are located in the brain. Ten years later, Chamot and El-Dinary (2000) proposed that not only are strategies mental procedures that assist learning but they also include overt activities. This being the case, a search for a denition has to encompass a spectrum of human activity ranging, for example, from visual imaging, as when memorizing vocabulary, to purposeful socializing, as when seeking out the opportunity to speak to someone in the L2. These broader denitions add to the problems regarding the location of strategies. The size-abstractness dilemma appears to be, in part, related to the action component in learner strategies. OMalley and Chamot (1990) provided a denition grounded in cognitive theory when they proposed that strategy applications resemble production systems (p. 52) with if and then clauses, for example:
IF the goal is to comprehend an oral or written text, and I am unable to identify a words meaning, THEN I will try to infer the meaning from context. IF the goal is to comprehend and remember an oral passage, and I have heard a complete passage or thought expressed, THEN I will summarize the passage to ensure I understand it. (p. 52)

323 Whereas some authors have tried to be allencompassing in their strategy denition, others have attempted to overcome the problem of size by claiming that although strategies are mental operations, they can move across categories. For example, Phakiti (2003) proposed that research should not differentiate between cognitive and metacognitive strategies; rather, it should identify the underlying goals or motivations for using a strategy and thereby dene a strategy as either cognitive or metacognitive. This procedure may be useful for reducing the total number of possible strategies, but it invites the abstractness criticism by creating two problems. First, how can the same mental operation be directly involved with language processing and also be responsible for monitoring that activity? Second, as I hope to show, goals and motivations are so integral to strategies that to change them changes the actual nature of the strategy itself. Cohen (1998) attempted to resolve the sizeabstractness problem by proposing a hierarchy of strategies. He proposed that improving reading comprehension be regarded as an overarching strategy and that use ongoing summaries written in the margin in telegraphic form be considered a substrategy (p. 10). Although this hierarchy helps to solve the problem of strategy size, it does not resolve the problem of abstractness. For example, does changing substrategies that are lower down in the hierarchy qualitatively change the nature of a strategy farther up? Additionally, this attempt at a solution in fact enlarges the meaning of a strategy even further by including intention to act as well as action itself. Improving reading comprehension to my mind, is not a recognisable distinct action but an abstracted determination to improve a skill. Use ongoing summaries written in the margin is a series of motor actions that can be observed as they are applied to a specic task but may relate to a whole series of mental actions. For example, the latter strategy, applied to preparing for an exam, could have as its goal not comprehension but memorization of material, which would link it to a completely different general determination of doing well on the exam. Thus to view strategies within hierarchical systems does not, in the end, provide us with a clearer construct. Moreover, it raises another issue: whether strategies are specic to a task or generalizable to all kinds of learning situations. Other authors have turned to stricter dictionary denitions of what a strategy is. The New Oxford Dictionary of English (1998) proposed that a strategy is a plan of action or policy designed to achieve a major or overall aim (p. 1837),

In these production systems, it is not clear which actions, mental or behavioural, and how many actions are involved in inferring meaning from context or in summarizing the passage. The nature of action in a strategy remains problematic because what strategies are purported to do is not transparently related to what strategies are.

324 whereas action is the fact or process of doing something (p. 17). McDonough (1995) therefore dened strategies as articulated plans for meeting particular types of problems not a piece of problem-solving itself (p. 3). Similarly, Phakiti (2003) dened strategies as not actually strategies in the strictest sense of the term. Rather, they should be seen as learners stable long-term knowledge of their strategy use (p. 681). These denitions would rule out action as an integral component of a strategy, invalidating the production systems referred to previously. According to these denitions sounding out an unfamiliar (written) word in order to see if anything comes to mind would not be a strategy. In which case, what could it possibly be? The literature reveals a further problem with strategy denition: the interchangeability of many of the terms used. For example, Oxford (1990) noted that strategy like tactic implies planning, competition, conscious manipulation, and movement toward a goal (p. 8). She proposes that strategies are a plan, step, or conscious action towards achievement of an objective (p. 8). Rubin (1987) also described strategies as any set of operations, plans, or routines used by learners to facilitate the obtaining, retrieval, storage and use of information (p. 19). Alexander, Graham, and Harris (1998), reviewing strategy research in the general educational literature, showed a concern for this interchangeable nature of terms used but were only able to propose a distinction between skills and other human action (mental or behavioural), not a distinction between strategies and other terms used: skills are procedures [italics added] that have been routinized. That is, students hone these techniques [italics added] to a level of automaticity, enabling them to perform a given task uidly and effectively (p. 135). Thus, a further unresolved problem is the semanticequivalence dilemma, with words like strategy, operation, routine, process, procedure, action, tactic, technique, plan, and step, being interchangeable in the literature. Few authors have tackled the problem of avoiding semantic equivalence and, when they have done so, the solution provided is not a complete one. McDonough (1995), for example, linked semantic equivalence to an appraisal of the differences among strategies, processes, and skills. He dened skills as performance-related, applicationspecic, subject to different levels of achievement, and, usually, leading to success. In other words, skills are operating at the level of overt behaviour even though they may reect all kinds of mental activity occurring below the surface. Processes are dened by McDonough as mechanisms through

The Modern Language Journal 90 (2006) which a set of information is transformed (p. 3). Providing these distinctions (between skills and processes), as I hope to demonstrate, is a useful line of reasoning to pursue. However, as noted earlier, McDonough did not recognise the action component of a strategy and it is therefore left unresolved how strategies might relate to processes and skills. Other researchers have tried to resolve the ambiguity with regard to what learner strategies are for or what they do. Mayer (1988) referred to strategies as behaviours of a learner that are intended to inuence how the learner processes information (p. 11). Strategies in his sense, therefore, appear to be additional to the processing of information. A number of authors also described the effects of strategy use in terms of facilitating learning or making learning more effective (e.g., Oxford, 1990; Vandergrift, 1998) rather than as part of the process itself. However, the link between the deployment of strategies in a task and eventual learning is not clear. For example, Vandergrift described strategies in listening comprehension as the strategies used by L2 listeners to facilitate their comprehension of oral texts (1998, p. 370). This description raises two questions. First, what is the relationship between something that facilitates listening comprehension and the development in the skill of listening? Certainly the way a strategy (or combination of strategies) leads to better understanding of an oral text in a specic task has been well documented (Chien, 1998; Harley, 2000; Laviosa, 2000; OMalley et al., 1989; Vandergrift, 1998). However, documenting how a strategys deployment leads to successful listening generally is more contentious, as Tsui and Fullilove (1998) discovered. Second, if strategies are the actions that facilitate comprehension, what are all the other actions that L2 listeners deploy that do not facilitate comprehension? Are all the former always effective? Are the latter all ineffective or redundant? There are few examples in the literature of individual ineffective strategies (rather than inappropriate combinations, viz., Vann & Abraham, 1990), and when they are proposed (e.g., Mitchell, 1992), they are described in such negative terms that they seem merely to be the opposite of any action with intention to learn: skimming over a communication, overlooking parts of a communication, ceasing work on a task in the belief that it is nished, blind application of memorized procedure, lack of any strategy to cope with getting stuck (p. 63). So, to return to our listening example, is it the case that all actions are strategies but that some actions facilitate listening comprehension and some do not? Or is it the case that in some situations

Ernesto Macaro strategies facilitate comprehension, and in some situations they do not? One answer to these questions would nd itself tracing a line of argument back to the notion of the good language learner (Naiman et al., 1978/1996). However, we are still not clear whether the good language learner only deploys effective strategies, or whether the good language learner deploys strategies effectively. In other words, does the good language learner select only those strategies that he or she has come to realize are effective, or does the good language learner orchestrate combinations of strategies effectively, regardless of their status as effective or otherwise? There remain, therefore, a number of unresolved issues and questions that undermine the theoretical basis of learner strategy research. As Rees-Miller (1993) argued, until researchers have resolved these problems, it is unlikely that many teachers will risk giving up valuable classroom time in order to train students to use strategies. These problems all relate to the lack of theory underlying the construct labelled a learner strategy , or to the lack of consensus as to the unit of analysis for learner strategy research. The problems can be summarised as follows:

325 There is clearly a need to revise the theoretical underpinnings of learner strategy research. However, the conclusions from the previous two sections should be viewed within a historical context. The work carried out by researchers in the eld of learner strategy was motivated by a genuine desire to alter the emphasis of research endeavour by examining language acquisition from the learners perspective, by looking at the process of learning rather than at the product of instruction, and by bridging the gap between theory and practice. It is particularly this last motivation that, in my view, led authors to posit strategies in broad terms that language teachers could more easily implement with their students. It is not, therefore, the purpose of this article to deride previous work on learner strategies but to attenuate some of the problems posed by these dilemmas. In order to achieve this purpose, I will propose a framework within cognitive psychology and information processing, one that includes learner strategies in a clear relationship with other domains of language learning and language use. Within this framework (see Figure 1), strategies are described as having a series of essential features rather than dened, in order to avoid semantic interchangeability and circularity of argument. There are three underlying propositions to the learner strategy features. It may help the reader if these are stated at the outset. 1. Researchers should describe strategies in terms of a goal, a situation, and a mental action. 2. Strategies are the raw material of conscious cognitive processing, and their effectiveness or noneffectiveness derives from the way they are used and combined in tasks and processes. 3. Strategies need to be distinguished from subconscious activity, language learning processes, skills, learning plans, and learning styles. RELATIONSHIP BETWEEN SUBCONSCIOUS ACTIVITY AND LEARNER STRATEGIES First, the framework argues for a distinction between learner strategies and subconscious activity in the brain (Figure 1). The basis for this proposition is as follows. Subconscious activity interacts with neurological processes and results in changes over which the language learner or user exerts virtually no conscious control. Examples of these activations, processes, and changes are activation of nodes in the brain, activation of propositional networks, abstraction of linguistic information (Norman & Rumelhart, 1975), inhibition

1. There is no apparent consensus about where learner strategies occur, inside the brain or outside it. 2. There is no consensus about what learner strategies are. Do they consist of knowledge, intention, action, or all three? 3. It is unclear how general or abstract learner strategies are and whether there exist substrategies as well as strategies and, as a consequence, if they can be classied in a framework or a hierarchy. 4. A lack of clarity also exists about whether their integrity survives across learning situations, tasks, and contexts. 5. There is no consensus about what they do, especially whether they are always facilitative and effective. 6. It is unclear whether they are integral to language processing or if they are some kind of extra facility that speeds up learning. 7. Strategy denition in the literature is arrived at through the use of equally undened terms. 8. There is a lack of consensus on a strategys relationship to skills and processes. 9. A lack of consensus remains on how strategies lead to both language learning and skill development over the long term.

326

The Modern Language Journal 90 (2006)

Strategic Plans

L2 Skills
h
Measurable and observable

L2 Processes
Interaction of strategy clusters as applied to language tasks

Cognitive Styles

Learning Styles

Learner Strategies
c
In working memory

g d

Motivation

Cluster 1

Cluster 2

Subconscious Activity

FIGURE 1 A Cognitive Framework for Learner Strategies

From proceduralization of declarative knowledge From implicit learning

of neural pathways (Baddeley, 1997), creation of logogens through multiple modalities (e.g., representational units of the verbal system and pools of strong connections in long-term memory; Eysenck, 1994), lexical storage (Baddeley, 1986), language-specic tagging of lexical items (Libben, 2000), and restructuring of mental models of the rule system at a biological or representational level (McLaughlin, 1990). These processes may at one time have been affected by conscious strategy deployment in working memory (and indeed may be once again), but in long-term memory they are largely not under the control of the individual (Cowan, 1999).2 Berry and Dienes (1993) proposed a notion of a threshold of cognition below which learning can be considered implicit. Although there is no precise distinction between explicit and implicit learning, and although it is as yet unclear how working memory deals with subconscious activity and its role in activating proceduralized knowledge, there is a growing consensus that working memory, focal awareness, attention, control, and consciousness are, at least, complementary concepts if not essentially the same (N. Ellis, 2001; Schmidt, 1990). Implicit learning, on the other hand, is characteriszed by a relative inaccessibility of knowledge via free recall, a sense of intuition, and limited transfer to related tasks (Berry & Dienes, 1993). Whereas subconscious activity is not under the

direct control of the individual, access to longterm memory is, in part, through the gateway of working memory (see arrow a in Figure 1). Proceduralization of declarative knowledge takes place through this gateway. Although total consensus does not exist as to the distinction between explicit and implicit learning (Hulstijn, 2005), by controlling (inter alia) for degree of awareness, task time available, and systematicity, we are beginning to create the circumstances where in the future that distinction will be possible (R. Ellis, 2005), and we are already observing certain phenomena that can only be attributed to implicit learning (Tokowicz & MacWhinney, 2005). This kind of learning would have to bypass working memory, at least in part, and be the result of some kind of direct contact with a language process, in our case an L2 process (see arrow b in Figure 1). I will return to L2 processes in the section entitled Relationship Between Learner Strategies and Second Language Processes. LEARNER STRATEGIES: FEATURES AND COMBINATIONS If we accept that working memory interacts with subconscious activity in long-term memory, then strategic action, I propose, must have, at the very least, its origins in working memory.3 The issue in the past has been whether learner strategies can span other components elsewhere in the

Ernesto Macaro framework and still be considered strategies. In order to resolve this issue, the framework proposes that learner strategies occur only in working memory and that they become other constructs elsewhere. For this proposition to hold true, a number of features are required in order to identify, describe, or constitute a learner strategy. Location of Strategies The location of strategies is the rst feature. Given that the framework proposes that learner strategies are located in working memory, they are subject to that components nite number of processing resources. In order to manage these resources, the central executive (Baddeley, 1997) has to exert control over their deployment. Conversely, without learner strategies, working memory could not carry out the perceiving, holding, processing, and encoding functions that it is required to do. Note that so far I have not made a distinction between general learner strategies and language-specic learner strategies. Strategies, then, are to be classied as conscious mental activity (Kail & Bisanz, 1982). Rabinowitz and Chi (1987) argued that mental processes that are deployed without awareness cannot be considered strategic. This assertion is supported by the proposition that strategies must contain not only an action but also a goal and a learning situation. Flavell, Miller, and Miller (1993) contrasted human memory with equine memory to argue this point. Although no psychologist would doubt that horses retain some memories, humans have the memory capacity not only for intentional and planful strategies, but also for evaluating knowledge about strategies. Thus, whereas a mental action might be subconscious, an action undertaken with a goal and evaluated against a learning situation can only be conscious (but see the Strategies and Tasks section on the issue of automaticity). Size, Abstractness, and Relationship to Other Strategies The size of a strategy is its second feature, as is the potential for it to be described, and its relationship to other strategies within a given task. In order to avoid the dilemma of size and abstractness, a strategys description should be effected at the lowest relevant level of articulation within the boundaries of conscious cognition. In other words, it should not be possible to describe a strategy by referring to a number of relevant subordinate strategies. Given that the

327 framework situates strategies in the domain of cognitive behaviour, not overt motor behaviour, a strategy is more appropriately described in terms of thinking rather than doing. By describing the mental action component of a strategy, we can begin to understand what the strategy actually purports to do. If we do this, we soon discover that strategies such as rehearsing and memorizing or rereading texts 4 are in fact combinations of strategies. For example, the act of rereading a text one has just written with the intention of improving it may involve strategies such as does it sound right? does it look right? what are the mistakes I usually make? and so on. Note that does it sound right? and does it look right? involve the phonological and visual components of working memory. Recalling mistakes usually made involves the interaction of working memory and long-term memory. Developmental psychologists have discovered that older children do not necessarily possess more cognitive strategies than do younger children but they become more exible and efcient in how they invest their resources (Flavell et al., 1993, p. 254). Thus, for a strategy to be effective in promoting learning or improved performance, it must be combined with other strategies either simultaneously or in sequence, thus forming strategy clusters. Effective learners deploy strategies in clusters appropriate to contexts and tasks (Macaro, 2001, 2003). Many strategy elicitation studies now recognise the clustering effects of strategies, for example, in L2 to L1 dictionary tasks (Neubach & Cohen, 1988), in listening tasks (Laviosa, 2000), and in reading tasks (Graham, 1997). An example of a strategy cluster can be found in how learners look up a new L2 word in an L1L2 dictionary when writing. Here such strategies as the following may be deployed: remember prior problems with dictionary use ; predict what problems I might encounter this time ; think about what part of speech I am looking for ; compare all definitions given ; compare collocations in L2 and L1; evaluate predictions ; remember to copy word correctly; check that it makes sense in the sentence generated. This cluster might, in turn, be combined with another cluster of strategies (see cluster representation in Figure 1) pertaining to memorising the new word for future use. Tasks impose differential demands on memory resources, and psychology research has tried to uncover the nature of processing resources and the principles that map task proles on to resource demands (Eysenck, 1994, p. 114). Strategy clusters may be ineffective if individual strategies within them are inappropriately

328 orchestrated. For example, Porte (1995) found that a subvocalizing strategy was counterproductive in a sentence copying task for some subjects. As the subjects moved from looking at text on a computer screen (which then disappeared) to writing the text down on paper, they subvocalized the text thereby losing the accuracy of the graphic form through (among other things) L1 interference. Macaro (2001) found that, among some young learners, applying personal schemata (prior knowledge of a topic) was ineffective in a reading task, because they did not combine and evaluate this prior knowledge with evidence available later in the text. There is some evidence, therefore, that orchestration of clusters of strategies, that is, choosing and evaluating from a range of strategies, is more effective than linear deployment of several strategies. Orchestration therefore suggests higher levels of metacognition. Effective strategy clusters, therefore, include and are evaluated via a metacognitive strategy or series of metacognitive strategies. Metacognitive strategies are contained in strategy clusters in order to regulate conscious cognitive activity (Schraw & Moshman, 1995). In the dictionary cluster previously described, metacognition would monitor and evaluate the cognitive strategies being deployed. Thus a classication of strategies that retains its explicatory power is cognitive as opposed to metacognitive (OMalley & Chamot, 1990), and it is retained in this framework. Strategies are either directly involved in working memory processing (perception, decoding, processing, storage, and retrieval) or they oversee cognitive strategies via planning, monitoring, and evaluating for effectiveness. Working memory resources are either occupied directly with language by operationalizing cognition or less directly by standing back and evaluating via metacognition. I want to propose that metacognitive strategies subsume affective strategies (recognised as a different category by OMalley & Chamot, 1990) because the latter require knowledge of oneself as a learner through recurrent monitoring of ones learning. Affective strategies, therefore, are part of the recursive use of metacognition to evaluate past cognitive strategies in culturally situated learning situations. I also propose that social strategies are clusters of cognitive and metacognitive strategies that lead to Strategic Plans (see arrow d in Figure 1). If a student of an L2 seeks out interaction with native speakers of that language in order to improve his or her learning, perhaps overcoming fear and shyness, he or she is not, in effect, doing anything other than deciding on a plan of action based on

The Modern Language Journal 90 (2006) a cluster of strategies previously evaluated. I will return to strategic plans in the Strategic Plans and Motivation section of the article. Strategies and Goals A mental action, I have argued, is a necessary component of a strategy, although not a sufcient one. The third feature of a strategy is its incorporation of a goal, and a strategys description is enhanced through the specication of a goal or goals. This proposition has substantial theoretical and empirical evidence in the literature of the psychology of motivation (see D ornyei, 2001, or Weiner, 1992, for a review). Human action is normally considered to be directed by purpose and dependent on the pursuance of goals. According to Locke (1996), for goals to be effective motivators for action, they (a) have to be established through the free choice and commitment of the individual, (b) must be specic and explicit, and (c) have to appear attainable. Strategies (within clusters of strategies) can be deployed in order to satisfy a teachers requirements, or they can be deployed in order to satisfy the learners own learning goals (Entwhistle, 1988). The two obligations are not mutually exclusive, but they can conict in certain learning environments. Erler (2003) found that young learners of French were deploying a number of strategies related to reading merely to satisfy the task requirements set by the teacher rather than to get the most out of a text. Learners themselves may not have clear ultimate goals. It is not for the learner to specify his or her goal (although it helps) but for the researcher or theoretician to infer or articulate it. Therefore, a key feature of a strategy should be the explicitness of its goal orientation. This explicitness is all the more needed given that strategies are often spoken about in the same context as self-determination and self-regulated learning (Dickinson, 1988; D ornyei, 2005; Pintrich, 1995; Wenden, 1995). Strategies and Tasks The fourth feature of learner strategies is that they are both situation-specic and transferable to other situations or tasks.5 Their transferability needs to be articulated by researchers, or at least this articulation should be one goal of learner strategy research. Without this transferability, the potential of learner strategies for learning is diminished. On the one hand, a learner needs to be able to apply a strategy consciously to an L2 process such as memorization (Oxford, 1990; Nunan,

Ernesto Macaro 1997), thereby strengthening the metacognitive link between the strategy and the achievement of recall. In this respect, the power of a strategy is its suitability to a particular task (Rabinowitz & Chi, 1987, p. 84). On the other hand, the strategy attains greater robustness if it contributes to a parsimonious framework that can be applied to a number of learning situations (the economy principle; see R. Ellis, 1985; Kellerman & Bialystok, 1997). As J. R. Anderson (1980/2000) proposed, it is likely that strategies are transferred to similar tasks by a procedure involving pattern matching through which the learner perceives similarities between the new task and former tasks where strategies were applied. However, I would argue on the basis of additional considerations to be described, that a pattern-matching procedure is not sufcient. Evaluation of strategy effectiveness, by learners, is likely to be undertaken against a background of the relative effectiveness of strategy clusters. Thus (taking into account the four features described), a strategy must conform to the algorithm: if in a learning situation/task X, and when the learning goal is Y, then try mental action Z. These then, are the main features of learner strategies. I will now discuss other considerations involved in the description of learner strategies. A strategys clarity is enhanced if its potential for leading to learning is articulated by researchers, even if only at the level of hypothesis. Putting a word into a sentence so as to remember it is one strategy cited by several authors when they discussed vocabulary-related strategy instruction (Grenfell & Harris, 1999; Nation, 2001; Oxford, 1990). But how does putting the word in a sentence help the learner remember it? Some strategies lend themselves to clear opportunities for hypothesis-testing in their interaction with subconscious activity. For example, the theoretical justication underlying the keyword strategy (Avila & Sadoski, 1996; Beaton et al., 1995; Lawson & Hogben, 1998) is provided by a rational account of how a link is produced between a new L2 word and an L1 word via the generation of a keyword that is a combination of sound and image. Given that retrieval of the L2 word operates in reverse order, the interface between an explicit strategy and an implicit process is explained logically. Thus, researchers are at least in a position to predict the strategys effect on long-term memory. Similarly, in a language-use situation, we are given insights into how strategies can be deployed in order to lighten the processing load on working memory. An example is the strategy of using the L1 for inner-speech when reading an L2 text in order to maintain concentration (Kern, 1994).

329 In different learners, a strategy will be at different levels of automaticity (see McLaughlin, 1987, and DeKeyser, 2001, for a review) or proceduralization ( J. R. Anderson, 1980/2000). Automatization must occur in order to speed up language learning and language use and to free up working memory space. Strategies have the potential, however, to be brought back to controlled attention for purposes of modication and adaptation (Kail & Bisanz, 1982), especially by adults who have the cognitive maturity to do so (Flavell et al., 1993). Control and automaticity relate to a strategys transferability. It may be that, through repeated practice and conrmation of effectiveness, a particular action Z becomes automatic in learning situation X. In this situation, three factors may require the strategy to be brought back into selective attention. First, the learning goal may change. For example, a student may embark on a course of academic writing, having previously learned the L2 in a speaking- and listening-based course. Clearly, for this student, a strategy such as avoid thinking in L1 will need to be reevaluated against the new learning goals. The student will have to evaluate whether academic writing is best achieved by avoidance of L1 mental resources. Second, the learning outcomes may appear unsatisfactory to the learner or to the teacher, and the learner will therefore need to bring back the strategy into selective attention each time situation X and learning goal Y are applicable in order for it to be reevaluated. Third, a strategy will need to be brought back into selective attention when a different learning situation presents itself, even if the goal has not changed, and the learner will need to evaluate the transferability of the strategy. A strategy may be difcult for some learners to deploy. An L1 speaker of English, for example, cannot easily deploy the keyword strategy when trying to memorise a Chinese word because of lack of phonemic correspondence. During some language processes, being below a certain linguistic threshold short-circuits the deployment of a strategy (Clarke, 1979; Lee & Schallert, 1997; Taillefer, 1996), as in the case of the transferability of L1 reading strategies to the comprehension of L2 written texts. There is other evidence of strategic deciencies (Rabinowitz & Chi, 1987). If, as has been suggested by a number of L1 studies, there is an association between variability in workingmemory limitations and nonword-repetition skills (Adams & Gathercole, 2000; Montgomery, 1995), and between working-memory limitations and L2 vocabulary acquisition (Gathercole & Baddeley, 1993; Service, 1992), then it is possible that

330 clusters of strategies are difcult for some learners to deploy in particular tasks. Clusters of strategies interact with L2 processes (see arrow c in Figure 1). The psychology literature demonstrates that strategies are not unique to L2 learning. Memorising a sequence of seven digits is not intrinsically different from memorising the names of seven food items in the L2. However, it is in the application of strategies to L2 tasks that (explicit) L2 processes take place. It is to these that we now turn.

The Modern Language Journal 90 (2006) integration into existing schemata. Each state provides the L2 user with some operational power. Written formulation (Bereiter & Scardamalia, 1987; De Larios et al., 1999; Flower & Hayes, 1981; Macaro, 2003) is a stage-type process in that each stage has no operational power on its own. It involves passing from an ideational stage to a stage where ideas begin to have linguistic form before moving on to further necessary stages (e.g., monitoring and evaluating). Strategies involved in the process of formulation are likely to include attempts to retrieve language chunks, evaluations of these chunks, attempts to restructure these chunks (change function words within them), word-forword translations at the phrasal level, and decisions to avoid using certain phrases. The framework proposes that repeated activation of language processes in working memory results in structural changes taking place in longterm memory both in vocabulary and morphosyntax (arrows c to a in Figure 1). These changes, together with repeated activation and automatisation of processes, lead to skill development. I will return to language skills in the section entitled Language Skills. STRATEGIC PLANS AND MOTIVATION Learner strategies also interact, via motivation, with strategic plans (arrows d and e in Figure 1). This notion of plans , mentioned previously, needs further exploration. Strategic plans have much broader learning objectives than do strategies. Plans, I want to argue, are constructed through an individual learners metacognitive theories. According to Schraw and Moshman (1995), metacognitive theories integrate metacognitive knowledge and experiences, and permit the learners own explanation and prediction of cognitive behaviour. However, according to the sources of the metacognitive theory (cultural, individual, or peer-constructed) and the theory types (tacit, explicit but informal, explicit and formal), metacognitive theories will vary in their effectiveness when implemented. Metacognitive theories, then, have a strong afnity with motivation, as dened by recent and more complex models (D ornyei, 2001; Weiner, 1992). I have already alluded to goal theory in arguing that the presence of a goal is a necessary condition for the construct of a strategy. Goals are, of course, also recognised components of motivation. Another important component of motivation is attribution , that to which learners attribute past success or failure (Williams & Burden, 1997). Whereas metacognitive theories predict future cognitive behaviour, attribution of

RELATIONSHIP BETWEEN LEARNER STRATEGIES AND SECOND LANGUAGE PROCESSES Language processes, and in our case L2 processes, consist of clusters of cognitive and metacognitive strategies in interaction with one another (see Figure 1). Language processes are therefore conscious although, like the learner strategies they comprise, they may operate so quickly as to appear subconscious. Processes bring about development involving a number of changes and involve a dynamic sequence of different stages (Faerch & Kasper, 1983, p. 30). They therefore involve transformation of language from one state to another or from one stage to another or both. The difference between state-like and stage-like processes is that, in the former, intermediate states of language transformation are usable by the L2 learner, whereas all the stages in a transformation need to be achieved for certain tasks to be completed. I will take two examples of language processes in order to try to illustrate these points. Reading in the L2 involves two interactive statelike processes: top-down and bottom-up processes (Grabe & Stoller, 2002; Kember & Gow, 1994; Stanovich, 1980). Each of these processes is made up of a cluster of strategies in a dynamic. Some strategies within this dynamic are applied in sequence. For example, Doctor and Klein (1992) detected an obligatory graphemephoneme check, followed by an optional spelling check in bottom-up processes in reading. Other strategies are applied more randomly according to task demands. Strategies interacting in these processes would include application of prior knowledge (Carrell, 1989), the application of common sense or logic, segmenting strings of linguistic units (Harrington, 2001), word-level and below word-level decoding, and use of adjacent text. These processes are used to attempt to transform L2 text from a state in which it is not understood into different states or levels of understanding, elaboration, and

Ernesto Macaro past success or failures will predict the way future strategic plans are actually formulated. The strength and endurance of the motivation, leading to the successful carrying out of a strategic plan, may well depend on another component of motivation, self-efcacy (Bandura, 1993). Selfefcacy is the relative measure of what learners think they can do in a given taskthe how good am I at this? factor. Successful deployment of strategy clusters in relation to a task builds higher levels of self-efcacy, leading to enhanced motivation. It has, after all, been one of the fundamental claims of learner strategy research that effective strategy deployment leads to increased motivation. A strategic plan, then, may arise from clusters of strategies that have formed some kind of metacognitive theory. The plan will be driven and shaped by motivational components such as attribution and self-efcacy. As an example, a learners decision to develop greater condence when speaking the L2 in large groups would trigger a strategic plan. This decision may have arisen from the metacognitive evaluation of previous experiences in similar situations where the learner may have deemed his or her linguistic behaviour inadequate. A cluster of communication strategies (e.g., deciding to use llers to buy processing time, chunking together longer speech runs between pauses, deciding to trigger comprehension responses from interlocutors ) would be deployed when future social situations occurred and the strategies would be evaluated for their effectiveness. LEARNING STYLES AND COGNITIVE STYLES Schmeck (1988) observed that if an individual has an inclination to use the same strategy in various situations, we can suspect the presence of a style (p. 8). It is unclear, from his description, what the causal factors of that style are or at what level (neurological, subconscious, consciouscognitive, behavioural) that style is operationalised. The framework in this article proposes that, if a cluster of strategies repeatedly interacts within an L2 process, that process may have an effect on cognitive styles (arrow f in Figure 1). Cognitive styles are the result of a habitual way of processing information (Das, 1988; Skehan, 1991) or, put another way, a habitual clustering of a selection of strategies in a given learning situation or task. Entwhistle (1988) described learning styles in terms of broad approaches or orientations to learning that are linked to motivational impulses. I interpret these approaches or orientations as clusters of cognitive

331 and metacognitive strategies that, having gone through a process of proceduralisation, have become stable and xed, and that a learner has a predisposition to use. They do not reect a deeper reality, a cognitive structure that is immutable as Hoffman (1997) suggested, let alone an even deeper neurological reality that governs cognition and therefore cognitive style. In the framework, therefore, cognitive styles combine with motivation (arrow g in Figure 1) to arrive at general learning styles that go beyond the cognitive level as in, for example, surface approaches to learning (Entwhistle, 1988). LANGUAGE SKILLS Language skills are the ability to carry out a language task with relative expertise to a relatively successful degree. These skills are summative-level manifestations (products) of L2 processes (arrow h in Figure 1) and are measurable. Skills can be measured either singly (listening, reading, speaking, writing) or in combinations as mixed skills (translations into the L2, summaries, reports). Skills can be measured both in terms of success at task achievement (as specied by a given set of criteria) and as rate of skill acquisition over time (against some kind of standardized norm). The automatisation of strategies, through the continual deployment of clusters of strategies during L2 processes, leads to the development of skillful behaviour. In the eld of L2 acquisition, as in the eld of experimental psychology, skills increase their efciency the more their underlying cognitive processes become proceduralised (J. R. Anderson, 1980/2000) and as reorganization into new and more efcient units takes place (McLaughlin, 1990). Let us take the case of speaking uency as an example. Towell, Hawkins, and Bazergui (1996), basing their research on the Levelt (1989) theoretical model of speaking, found that an increase in L2 uency after a year in the target country could be attributed primarily to the formulator whereby the Mean Length of Run between pauses meant that the participants were speaking faster because they were producing longer strings without pausing. The ability to say longer things without pausing is a surface manifestation of the practice effect of language processes without which chunking could not take place and whereby working memory limitations could not be overcome. DISCUSSION The framework proposed in this article is a response to a lack of theoretical rigour identied

332 in the research literature on learner strategies. It proposes a number of distinct components that are nevertheless in strong interaction with one another. In this way, it seeks to provide a number of advances in the eld. First, the framework rejects the dual location of a strategy and proposes that (a) a strategy occurs in the brain, in working memory; and (b) its description comprises a goal, a situation, and a mental action. In so doing, the framework attenuates the problems related to strategy size and strategy abstractness. Second, it describes strategies as integral components of processing theory rather than as shortcuts to faster processing. Strategies, with some exceptions, are available to all learners, operate in clusters, and in relation to language tasks become L2 processes. The nature of an individual strategy remains constant. It is the problems posed by task demands that vary and that bring about variation in the selection and orchestration of strategy clusters. Third, successful learning is no longer linked to the individual learners frequency of strategy use, but to his or her orchestration of strategies available to him or her. This framework will need to be tested empirically in order to see if previous ambiguities are indeed resolved. The features of strategies provided in this article offer a clearer construct that can be tested for its validity. In order to validate this framework, so that in future it might acquire the status of a model, future research needs to provide the following evidence: (a) that conscious cognitive activity can be described in terms of action, goal, and learning situation; (b) that strategy clusters can be systematically mapped against L2 tasks; (c) that automatised strategies can be brought back to selective attention and evaluated by learners; (d) that a reduction of categories, to cognitive strategies and metacognitive strategies only, is theoretically justiable and sufcient; and (e) that language learning and skills development result from the repeated successful activation of L2 processes, which are the relatively successful applications of strategy clusters to L2 tasks. I am currently beginning to explore item (b) through the following systematic research process: Step 1: verbal protocols carried out against a specic task; Step 2: validation and generalisation of a cluster in large groups of learners via taskbased self-report; and Step 3: factor analysis in order to measure change in cluster use in at least two time points, either through nonmanipulated development or through intervention (Macaro, 2005).

The Modern Language Journal 90 (2006) The framework addresses the problem of specicity and generalisability of strategies. In order to reduce the total number of possible strategies that could be attributed to the total number of learning situations or tasks, strategies must remain both situation specic and transferable. Future research could usefully establish how transferable a strategy is between two learning situationsfor example, by examining the application of prior knowledge to both reading and listening. The framework addresses the semanticequivalence problem by proposing that strategies are allocated a separate place within a framework, which also includes processes, skills, plans, and styles. Their interactions with subconscious processes have been described where that was possible from the available evidence. Their interaction with conscious L2 processes has been described in some detail and, although it is clearly an incomplete account of the possible interactions, the framework seeks to contribute to a theory of language acquisition situated in a connectionist paradigm operating in conjunction with a theory of human memory. In doing so, the framework clearly adheres to cognitive psychologys quest to demonstrate that human knowledge is organised and that learning needs to be interpreted in the light of this organisation (McLaughlin, 1990). Learning of the L2 is brought about, in long-term memory, via strategic behaviour in working memory, through the development of declarative and procedural knowledge. Performance in the L2 is enhanced as a result of the way clusters of strategies interact with language processes, and these in turn contribute to skills through their acceleration and automatisation. It is the way that clusters of strategies interact with processes and thence skills that both knowledge of the language and performance in the language progress. In that sense, strategies do not make learning more efcient; they are the raw material without which L2 learning cannot take place. More research is needed to illuminate how strategies combine to lead to both language competence and language performance.

CONCLUSIONS The author of this article recognises that an enormous amount of research effort has gone into the learner strategy area and that some progress has been made. Yet criticisms regarding lack of theoretical underpinnings have persisted for nearly 2 decades and are, to some extent, justied. However, the research contribution of

Ernesto Macaro authors, to whom these criticisms have been leveled, has been in providing a bridge between theory and practice and, therefore, this article is not meant as an attack on their work. Rather, I have attempted to provide a theoretical framework for that valuable research endeavour, one that predicts that language is developed through the interaction of the components in the framework and one that can be tested by empirical research. I anticipate that one of the biggest challenges to the proposed framework will come in the form of an accusation of strategy proliferation. By opting for a more irreducible approach to strategy description, we may well be looking at hundreds of individual strategies. Part of this challenge can be met by the notion of transferability and the economy principle discussed in this article. For the rest, it is my view that there is no other solution. It could mean a research agenda that takes an additional decade. To some, this may seem a daunting task not worth undertaking. However, without attempting to explain cognitive behaviour in all its complexity, as meticulously and scientically as possible, and then attempting to ascertain its relationship with overt behaviour on the one hand and subconscious behaviour on the other, we will continue to offer only a supercial glimpse into L2 learning from the learners perspective.
REFERENCES

333

ACKNOWLEDGMENTS I would like to thank Andrew D. Cohen, Victoria Murphy, Peter Gu, and the two MLJ anonymous reviewers for their invaluable comments and advice in the development of this article.

NOTES
1 These claims, as with the criticisms in the next section, are not necessarily claims being made by the current author. 2 The precise relationship between working memory and long-term memory is, however, still a matter for considerable debate (see Miyake & Shah, 1999). 3 See Miyake and Shah (1999) for an overview of different models of working memory. The one adopted here, however, is that of Baddeley (1997). 4 Example taken from OMalley and Chamot (1990). 5 Task, here, is meant both in the sense of target task and pedagogical task (Wenden, 1995).

Adams, A.M. & Gathercole, S. E. (2000). Limitations in working memory: Implications for language development. International Journal of Language and Communication Disorders, 35, 95116. Alexander, A., Graham, S., & Harris, K. R. (1998). A perspective on strategy research: Progress and prospects. Educational Psychology Review, 10, 129 154. Anderson, J. R. (2000). Cognitive psychology and its implications . New York: Worth Publishers. (Original work published 1980) Anderson, N. J. (1991). Individual differences in strategy use in second language reading and testing. Modern Language Journal, 75, 460472. Anderson, N. J., & Vandergrift, L. (1996). Increasing metacognitive awareness by using think-aloud protocols and other verbal report formats. In R. Oxford (Ed.), Language learning strategies around the world: Cross cultural perspectives (pp. 318; Tech. Rep. No. 13). Honolulu: Second Language Teaching and Curriculum Center, University of Hawaii. Avila, E., & Sadoski, M. (1996). Exploring new applications of the keyword method to acquire English vocabulary. Language Learning, 46, 379395. Aziz, L. (1995). A model of paired cognitive and metacognitive strategies: Its effect on second language grammar and writing performance . Unpublished doctoral dissertation, University of San Francisco, California. Bacon, S. M. (1992). The relationship between gender, comprehension, processing strategies and cognitive and affective response in foreign language listening. Modern Language Journal, 76, 160176. Baddeley, A. (1986). Working memory . Oxford, UK: Clarendon Press. Baddeley, A. (1997). Human memory: Theory and practice . Hove, UK: Psychology Press. Bandura, A. (1993). Perceived self-efcacy in cognitive development and functioning. Educational Psychologist, 28, 117148. Beaton, A., Gruneberg, M., & Ellis, N. (1995). Retention of foreign language vocabulary learned using the keyword method: A ten year follow up. Second Language Research, 11, 112120. Bereiter, C., & Scardamalia, M. (1987). The psychology of written composition . London: Erlbaum. Berry, C., & Dienes, Z. (1993). Implicit learning: Theoretical and empirical issues . Hillsdale, NJ: Erlbaum. Bialystok, E. (1981). The role of conscious strategies in second language prociency. Modern Language Journal, 65, 2435. Bialystok, E. (1983). Some factors in the selection and implementation of communication strategies. In C. Faerch & G. Kasper (Eds.), Strategies in interlanguage communication (pp. 100118). London: Longman. Block, E. (1986). The comprehension strategies of second language readers. TESOL Quarterly, 20, 463 493.

334
Bowles, M. A., & Leow, R. P. (2005). Reactivity and type of verbal report in SLA research methodology: Expanding the scope of investigation. Studies in Second Language Acquisition, 27, 415440. Bugel, K., & Buunk, B. P. (1996). Sex differences in foreign language text comprehension: The role of interests and prior knowledge. Modern Language Journal, 80, 1531. Burgos-Kohler, N. (1991) The effects of a selected group of language learning strategies upon language development . Unpublished doctoral dissertation, University of Texas at Austin. Carrell, P. L. (1989). Metacognitive awareness and second language reading. Modern Language Journal, 73, 121130. Carrell, P. L., Pharis, B. G., & Liberto, J. C. (1989). Metacognitive strategy training for ESL reading. TESOL Quarterly, 23, 647678. Chamot, A. U., & El-Dinary, P. B. (1999). Childrens learning strategies in language immersion classrooms. Modern Language Journal, 83, 319338. Chamot, A. U., & El-Dinary, P. B. (2000). Childrens learning strategies in language immersion classrooms . Washington, DC: National Capital Language Resource Center. Chien, C. (1998). The strategy use in listening comprehension for EFL learners in Taiwan. RELC Journal, 29, 6691. Christianson, K. (1997). Dictionary use by EFL writers: What really happens? Journal of Second Language Writing, 6, 2343. Clarke, M. A. (1979). Reading in Spanish and English: Evidence from adult ESL students. Language Learning, 29, 121150. Cohen, A. D. (1998). Strategies in learning and using a second language . London: Longman. Cohen, A. D., & Aphek, E. (1980). Retention of secondlanguage vocabulary over time: Investigating the role of mnemonic associations. System, 8, 221235. Cohen, A. D., & Hosenfeld, C. (1981). Some uses of mentalistic data in second language research. Language Learning, 31, 285313. Cohen, A. D., Weaver, S. J., & Li, T.-Y. (1995). The impact of strategies-based instruction on speaking a foreign language . Minneapolis: Center for Advanced Research on Language Acquisition, University of Minnesota. Conti, G. (2004). Metacognitive enhancement and error correction: An investigation in the impact of selfmonitoring strategies on L2 Italian student writing . Unpublished doctoral dissertation, University of Reading, United Kingdom. Cowan, N. (1999). An embedded-process model of working memory. In A. Miyake & P. Shah (Eds.), Models of working memory: Mechanisms of active maintenance and executive control (pp. 62101). Cambridge: Cambridge University Press. Das, J. P. (1988). Simultaneous-successive processing and planning: Implications for school learning. In R. R. Schmeck (Ed.), Learning strategies and learning styles (pp. 101129). London: Plenum Press.

The Modern Language Journal 90 (2006)

DeKeyser, R. (2001). Automaticity and automatization. In P. Robinson (Ed.), Cognition and second language instruction (pp. 125151). Cambridge: Cambridge University Press. De Larios, J. C., Murphy, L., & Manchon, R. (1999). The use of restructuring strategies in EFL writing: A study of Spanish learners of English as a foreign language. Journal of Second Language Writing, 8, 1344. Devine, J. (1984). ESL readers internalized models of the reading process. In J. Handsome, R. Orem, & B. Taylor (Eds.), On TESOL 83 (pp. 95108). Washington, DC: TESOL. Dickinson, L. (1988). Learner training. In A. Brookes & P. Grundy (Eds.), Individualisation and autonomy in language learning (pp. 4553). London: The British Council and Modern English Publications. Doctor, E. A., & Klein, D. (1992). Physical processing in bilingual word recognition. In R. J. Harris (Ed.), Cognitive processing in bilinguals (pp. 237252). Amsterdam: Elsevier Science. D ornyei, Z. (1995). On the teachability of communication strategies. TESOL Quarterly, 29, 585. D ornyei, Z. (2001). Teaching and researching motivation . London: Longman. D ornyei, Z. (2005). The psychology of the language learner . Mahwah, NJ: Erlbaum. D ornyei, Z., & Skehan, P. (2003). Individual differences in second language learning. In C. J. Doughty & M. H. Long (Eds.), The handbook of second language acquisition (pp. 589630). Oxford, UK: Blackwell. Ehrman, M., & Oxford, R. L. (1989). Effects of sex differences, career choice, and psychological type on adult language learning strategies. Modern Language Journal, 72, 253265. Ellis, N. (2001). Memory for language. In P. Robinson (Ed.), Cognition and second language instruction (pp. 3368). Cambridge: Cambridge University Press. Ellis, R. (1985). Sources of variability in interlanguage. Applied Linguistics, 6, 118131. Ellis, R. (2005). Measuring implicit and explicit knowledge of a second language: A psychometric study. Studies in Second Language Acquisition, 27, 141 172. Entwhistle, N. (1988). Motivational factors in students approaches to learning. In R. R. Schmeck (Ed.), Learning strategies and learning styles (pp. 2152). London: Plenum Press. Ericsson, K. A., & Simon, H. A. (1987). Verbal reports on thinking. In G. Faerch & G. Kasper (Eds.), Introspection in second language research (pp. 2453). Clevedon, UK: Multilingual Matters. Erler, L. (2003). Year 7 pupils experiences of reading French as a foreign language . Unpublished doctoral dissertation, University of Oxford, United Kingdom. Eysenck, M. W. (1994). The Blackwell dictionary of cognitive psychology . Oxford, UK: Blackwell. Faerch, C., & Kasper, G. (1983). Strategies in interlanguage communication . London: Longman.

Ernesto Macaro
Flaitz, J., & Feyten, C. (1996). A two phase study involving consciousness raising and strategy use for foreign language learners. In R. Oxford (Ed.), Language learning strategies around the world: Cross cultural perspectives (Tech. Rep. No. 13, pp. 157 166). Honolulu: Second Language Teaching and Curriculum Center, University of Hawaii. Flavell, J. H., Miller, P. H., & Miller, S. C. (1993). Cognitive development . Englewood Cliffs, NJ: Prentice Hall. Flower, L., & Hayes, J. R. (1981). A cognitive process theory of writing. College Composition and Communication, 32, 365387. Fraser, C. A. (1999). Lexical processing strategy use and vocabulary learning through reading. Studies in Second Language Acquisition, 21, 225241. Gathercole, S. E., & Baddeley, A. D. (1993). Working memory and language . Hove, UK: Psychology Press. Gillette, B. (1994). The role of learner goals in L2 success. In J. P. Lantolf & G. Appel (Eds.), Vygotskian approaches to second language research (pp. 195 213). Oxford: Oxford University Press. Grabe, W., & Stoller, F. L. (2002). Teaching and researching reading . London: Longman. Graham, S. (1997). Effective language learning . Clevedon, UK: Multilingual Matters. Grenfell, M., & Harris, V. (1999). Modern languages and learning strategies: In theory and practice . London: Routledge. Gu, Y. (2002). Gender, academic major, and vocabulary learning strategies of Chinese EFL learners. RELC Journal, 33, 3554. Gu, Y., & Johnson, K. J. (1996). Vocabulary learning strategies and language learning outcomes. Language Learning, 46, 643679. Harley, B. (2000). Listening strategies in ESL: Do age and L1 make a difference? TESOL Quarterly, 34, 769776. Harrington, M. (2001). Sentence processing. In P. Robinson (Ed.), Cognition and second language instruction (pp. 91124). Cambridge: Cambridge University Press. Hassan, X., Macaro, E., Mason, D., Nye, G., Smith, P., & Vanderplank, R. (2005). Strategy training in language learning: A systematic review of available research. In Research Evidence in Education Library. London: EPPICentre, Social Science Research Unit, Institute of Education, University of London. Retrieved from http://eppi.ioe.ac.uk/ EPPIWeb/home.aspx?page=/reel/review_groups/ m/review_one.htm Hoffman, S. Q. (1997). Field dependence/ independence in second language acquisition and implications for educators and instructional designers. Foreign Language Annals, 30, 222234. Holunga, S. (1994). The effect of metacognitive strategy training with verbalization on the oral accuracy of adult second language learners . Unpublished doctoral dissertation, University of Toronto, Canada. Hulstijn, J. H. (2005). Theoretical and empirical issues in the study of implicit and explicit second-

335
language learning: Introduction. Studies in Second Language Acquisition, 27, 129140. Kail, R. V., & Bisanz, J. (1982). Cognitive Strategies. In C. R. Puff (Ed.), Handbook of research methods in human memory and cognition (pp. 229255). New York: Academic Press. Kellerman, E. (1991). Compensatory strategies in second language research: A critique, a revision and some (non-) implications for the classroom. In R. Phillipson, E. Kellerman, L. Selinker, M. Sharwood-Smith, & M. Swain (Eds.), Foreign and second language pedagogy research (pp. 142162). Clevedon, UK: Multilingual Matters. Kellerman, E., & Bialystok, E. (1997). On psychological plausibility in the study of communication strategies. In G. Kasper & E. Kellerman (Eds.), Communication strategies: Psycholinguistic perspectives (pp. 3148). London: Longman. Kember, D., & Gow, L. (1994). An examination of the interactive model of ESL reading from the perspective approaches to studying. RELC Journal, 25, 125 Kern, R. G. (1994). The role of mental translation in second language reading. Studies in Second Language Acquisition, 16, 441461. Kim, D. (1999). An exploration of listening comprehension linked to authentic input and language learning strategies in a second language . Unpublished doctoral dissertation, University of Texas at Austin. Knight, S. (1994). Dictionary use while reading: The effects on comprehension and vocabulary acquisition for students of different verbal abilities. Modern Language Journal, 78, 285299. Koda, K. (1990). The use of L1 reading strategies in L2 reading: Effects of L1 orthographic structures on L2 phonological recoding strategies. Studies in Second Language Acquisition, 12, 393410. Kohler, D. B. (2002). The effects of metacognitive language learning strategy training on lower-achieving second language learners . Unpublished doctoral dissertation, Brigham Young University, Provo, UT. Lam, W., & Wong, J. (2000). The effects of strategy training on developing discussion skills in an ESL classroom. ELT Journal, 54, 245255. Laviosa, F. (2000). The listening comprehension processes and strategies of learners of Italian: A case study. Rassegna Italiana di Linguistica Applicata, 2, 129159. Lawson, M. J., & Hogben, D. (1998). Learning and recall of foreign language vocabulary: Effects of a keyword strategy for immediate and delayed recall. Learning and Instruction, 8, 179194. Lee, J.-W., & Schallert, D. L. (1997). The relative contribution of L2 language prociency and L1 reading ability to L2 reading performance: A test of the threshold hypothesis in an EFL context. TESOL Quarterly, 31, 713739. Leow, R. P., & Morgan-Short, K. (2004). To think aloud or not to think aloud: The issue of reactivity in SLA research methodology. Studies in Second Language Acquisition, 26, 3557.

336
Levelt, W. J. M. (1989). Speaking: From intention to articulation . Cambridge, MA: MIT Press. Levine, A., Reves, T., & Leaver, B. L. (1996). Relationship between language learning strategies and Israeli versus Russian cultural-educational factors. In R. Oxford (Ed.), Language learning strategies around the world: Cross cultural perspectives (Tech. Rep. No. 13, pp. 157166). Honolulu: Second Language Teaching and Curriculum Center, University of Hawaii. Libben, G. (2000). Representation and processing in the second language lexicon: The homogeneity hypothesis. In J. Archibald (Ed.), Second language acquisition and theory (pp. 229248). Oxford, UK: Blackwell. LoCastro, V. (1994). Learning strategies and learning environments. TESOL Quarterly, 28, 409414. LoCastro, V. (1995). The author responds. TESOL Quarterly, 29, 172174. Locke, E. A. (1996). Motivation through conscious goal setting. Applied and Preventive Psychology, 5, 117 124. Macaro, E. (2000). Learner strategies in foreign language learning: Cross national factors. Tuttitalia, 22, 918. Macaro, E. (2001). Learner strategies in second and foreign language classrooms . London: Continuum. Macaro, E. (2003). Teaching and learning a second language: A guide to current research and its applications . London: Continuum. Macaro, E. (2005, September). Reading strategies of young-beginner learners of French: An intervention study. Paper presented at the annual meeting of the British Association for Applied Linguistics, Bristol, United Kingdom. Mayer, R. E. (1988). Learning strategies: An overview. In C. E. Weinstein, E. T. Goetz, & P. A. Alexander (Eds.), Learning and study strategies: Issues in assessment, instruction, and evaluation (pp. 1122). San Diego, CA: Academic Press. McDonough, S. H. (1995). Strategy and skill in learning a foreign language . London: Arnold. McGruddy, R. (1999). The effect of listening comprehension strategy training with advanced-level ESL students . Unpublished doctoral dissertation, Georgetown University, Washington, DC. McLaughlin, B. (1987). Theories of second-language learning . London: Arnold. McLaughlin, B. (1990). Restructuring. Applied Linguistics, 11, 113128. Mitchell, I. (1992). The class level. In J. R. Baird & J. R. Northeld (Eds.), Learning from the PEEL experience (pp. 61104). Melbourne, Australia: Monash University Press. Miyake, A., & Shah, P. (1999). Models of working memory: Mechanisms of active maintenance and executive control . Cambridge: Cambridge University Press. Montgomery, J. W. (1995). Examination of phonological working memory in specically languageimpaired children. Applied Psycholinguistics, 16, 355378.

The Modern Language Journal 90 (2006)

Naiman, N., Frohlich, M., Stern, H., & Todesco, A. (1996). The good language learner . Clevedon, UK: Multilingual Matters. (Original work published 1978) Nakatani, Y. (2005). The effects of awareness-raising training on oral communication strategy use. Modern Language Journal, 89, 7691. Nation, I. S. P. (2001). Learning vocabulary in another language . Cambridge: Cambridge University Press. Nayak, N., Hansen, N., Krueger, N., & McLaughlin, B. (1990). Language learning strategies in monolingual and multilingual adults. Language Learning, 40, 221244. Neubach, A., & Cohen, A. D. (1988). Processing strategies and problems encountered in the use of dictionaries. Dictionaries: Journal of the dictionary society of North America, 10, 119. Norman, D. A., & Rumelhart, D. E. (1975). Explorations in cognition . San Francisco, CA: W. H. Freeman & Co. Nunan, D. (1997). Strategy training in the language classroom: An empirical investigation. RELC Journal, 28, 5681. OMalley, J. M. (1987). The effects of training in the use of learning strategies on acquiring English as a second language. In A. Wenden & J. Rubin (Eds.), Learner strategies in language learning (pp. 133144). Englewood Cliffs, NJ: Prentice Hall International. OMalley, J. M., & Chamot, A. U. (1990). Learning strategies in second language acquisition . Cambridge: Cambridge University Press. OMalley, J. M., Chamot, A. U., & Kupper, L. (1989). Listening comprehension strategies in second language acquisition. Applied Linguistics, 10, 418437. Oxford, R. L. (1990). Language learning strategies: What every teacher should know . Boston: Heinle. Oxford, R. L., & Burry-Stock, J. A. (1995). Assessing the use of language learning strategies worldwide with the ESL/EFL version of the strategy inventory for language learning (SILL). System, 23, 123. Oxford, R. L., & Nyikos, M. (1989). Variables affecting choice of language learning strategies by university students. Modern Language Journal, 73, 291300. Ozeki, N. (2000). Listening strategy instruction for female EFL college students in Japan . Unpublished doctoral dissertation, Indiana University of Pennsylvania. Phakiti, A. (2003). A closer look at gender and strategy use in L2 reading. Language Learning, 53, 649 702. Pintrich, P. R. (1995). Understanding self-regulated learning. In P. R. Pintrich (Ed.), Understanding selfregulated learning (pp. 312). San Francisco, CA: Jossey-Bass. Porte, G. K. (1995). Writing wrongs: Copying as a strategy for underachieving EFL writers. ELT Journal, 49, 144151. Porte, G. K. (1997). The etiology of poor second language writing: The inuence of perceived teacher preferences on second language revision strategies. Journal of Second Language Writing, 6, 6178.

Ernesto Macaro
Rabinowitz, M., & Chi, M. (1987). An interactive model of strategic processing. In S. J. Ceci (Ed.), Handbook of cognitive, social, and neuropsychological aspects of learning disabilities (Vol. 2, pp. 83102). Hillsdale, NJ: Erlbaum. Raymond, P. M. (1993). The effects of structure strategy training on the recall of expository prose for university students reading French as a second language. Modern Language Journal, 77, 445458. Rees-Miller, J. (1993). A critical appraisal of learner training: Theoretical bases and teaching implications. TESOL Quarterly, 27, 679689. Rubin, J. (1987). Learner strategies: Theoretical assumptions, research history and typology. In A. Wenden & J. Rubin (Eds.), Learner strategies and language learning (pp. 1529). Englewood Cliffs, NJ: Prentice Hall. Sanaoui, R. (1995). Adult learners approaches to learning vocabulary in second languages. Modern Language Journal, 79, 1528. Sarig, G. (1987). High-level reading in the rst and in the foreign language: Some comparative process data. In J. Devine, P. L. Carrell, & D. E. Eskey (Eds.), Research in reading English as a second language (pp. 105120). Washington, DC: TESOL. Schmeck, R. R. (Ed.). (1988). Learning strategies and learning styles . London: Plenum Press. Schmidt, R. W. (1990). The role of consciousness in second language learning. Applied Linguistics, 11, 129158. Schraw, G., & Moshman, D. (1995). Metacognitive theories. Educational Psychology Review, 7, 351371. Seliger, H. W. (1983). The language learner as linguist: Of metaphors and realities. Applied Linguistics, 4, 179191. Sengupta, S. (2000). An investigation into the effects of revision strategy instruction on L2 secondary school learners. System, 28, 97113. Seo, K. (2000). Intervening in tertiary students strategic listening in Japanese as a foreign language . Unpublished doctoral dissertation, Grifth University, Australia. Service, L. (1992). Phonology, working memory, and foreign language learning. Quarterly Journal of Experimental Psychology, 45a, 2150. Sheorey, R. (1999). An examination of language learning strategy use in the setting of an indigenized variety of English. System, 27, 173190. Skehan, P. (1989). Individual differences in secondlanguage learning . London: Arnold. Skehan, P. (1991). Individual differences in secondlanguage learning. Studies in Second Language Acquisition, 13, 275298. Stanovich, K. E. (1980). Toward an interactivecompensatory model of individual differences in

337
the development of reading uency. Reading Research Quarterly, 16, 3265. Stevick, E. W. (1990). Research on what? Some terminology. Modern Language Journal, 74, 143153. Taillefer, G. F. (1996). L2 reading ability: Further insight into the short-circuit hypothesis. Modern Language Journal, 80, 461477. The new Oxford dictionary of English . (1998). Oxford: Oxford University Press. Thompson, I., & Rubin, J. (1996). Can strategy instruction improve listening comprehension? Foreign Language Annals, 29, 331342. Tokowicz, N., & MacWhinney, B. (2005). Implicit and explicit measures of sensitivity to violations in second language grammar: An event-related potential investigation. Studies in Second Language Acquisition, 27, 173204. Towell, R., Hawkins, R., & Bazergui, N. (1996). The development of uency in advanced learners of French. Applied Linguistics, 17, 84119. Tsui, A. B. M., & Fullilove, J. (1998). Bottom-up and top-down processing as a discriminator of L2 listening performance. Applied Linguistics, 19, 432 451. Vandergrift, L. (1998). Successful and less successful listeners in French: What are the strategy differences? French Review, 71, 370394. van Hell, J. G., & Mahn, C. A. (1997). Keyword mnemonics versus rote rehearsal: Learning concrete and abstract foreign words by experience and inexperienced learners. Language Learning, 47, 507 546. Vann, J. R., & Abraham, R. G. (1990). Strategies of unsuccessful language learners. TESOL Quarterly, 24, 177198. Victori, M., & Lockhart, W. (1995). Enhancing metacognition in self-directed language learning. System, 23, 223234. Weiner, B. (1992) Human motivation: Metaphors, theories and research . London: Sage. Wenden, A. (1987a). Conceptual background and utility. In A. Wenden & J. Rubin (Eds.), Learner strategies in language learning (pp. 313). London: Prentice Hall. Wenden, A. (1987b). Incorporating learner training in the classroom. In A. Wenden & J. Rubin (Eds.), Learner strategies in language learning (pp. 3142). London: Prentice Hall. Wenden, A. (1995). Learner training in context: A knowledge-based approach. System, 23, 183194. Wenden, A., & Rubin, J. (Eds.). (1987). Learner strategies in language learning . London: Prentice Hall. Williams, M., & Burden, R. (1997). Psychology for language teachers . Cambridge: Cambridge University Press.