Despite the profusion of rich and varied research on the role of Learning Strategies (LS) in second language acquisition (SLA), today, some twenty years on from the first attempts at exploring how learners go about the task of learning a second language, researchers are still struggling to agree upon a universally accepted definition for the concept of strategy. In 1991 Seliger complained that the indiscriminate use of the word strategy in SLA literature had brought us to a point of 'semantic satiatiorí in which the term had become devoid of any real meaning (1991, p- 36). Less pessimistically, but in the same vein, Ellis (1994) referred to the concept of 'strategy' as 'fuzzy ... and not easy to tie dowrí (1994, p. 529).
The confusion surrounding the definition of 'strategy' arises from the research literature itself where the term has often been substituted for synonyms which have blurred the inherent meaning of the word by equating it with other similar concepts. Wenden (1987) has pointed out the multi-purpose use of the term to refer to all of the following: techniques, tactics, potentially conscious plans, consciously employed operations, problem-solving procedures, etc. (1987, p. 7). In fact a closer look at the definitions of Learning Strategies offered by different researchers indicates the high degree of overlapping and lack of precision in their choice of terminology. While for Naiman et al. (1978), Stern (1983), Schmeck (1988) and Seliger (1991) stratégies are seen as general learning approaches, with the more specific learner actions receiving the ñame of techniques or tactics, Rubin (1981) refers to general cognitive processes and specific stratégies. In her definition Wenden (1987) refers to language learning 'behaviours' while O'Malley and Chamot (1990) speak about 'thoughts or behaviours', leaving us in doubt as to whether strategies are to be considered as behavioural or mental or both. Finally Chesterfield and Chesterfield's (1985) definition reflects their concern with learner interaction while Oxford (1990) stresses the affective side of learning. Unable, then, to agree upon one generally accepted definition of Learning Strategies, researchers have had to resort to listing what seem to be their main characteristics in an attempt to solve this conceptual problem (Wenden, 1987; Ellis, 1994). This has clarified very little, however, as the characteristics cited tend to be contradictory and vague. For example, Ellis (1994) states that 'Strategies refer to both general approaches and specific actions or techniques to learn an L2' or that 'Some strategies are behavioural while others are mental. Thus some are directly observable while others are noí (1994, p. 532). Wenden (1987) lists problematicity (potential), consciousness and the directness/indirectness of their effect on learning as among their defining characteristics. The persistence of such ambiguity around the construct of strategy is detrimental to the concerns of empirical research in the fields of second and foreign language development.
Researchers have recognized the need to try to achieve some coherence across the field in terms of both the descriptive terminology and the conceptual characteristics inherent to the construct of strategy and while attempts in this direction have been made (Willing, 1989; Bialystok, 1990; Oxford and Cohen, 1992) as yet no consensus has been reached. Given the diverging opinions of individual researchers on a series of important, yet conflictive, issues relating to the definition of Learning Strategies, it is vital that any discussion of strategies, in whatever context, should begin with an explicit statement of the position adopted with respect to these crucial conceptual and classificatory problems.
The most serious points of contention in Language Learning Strategy research include: 1. The overlapping which occurs between terms such as process, strategy, tactic and...