Competence and performance
The use of ‘competence’ and ‘performance’ as technical terms originates with CHOMSKY (1965:4). ‘Competence’ refers to ‘the speaker-hearer’s knowledge of his language’ and ‘Performance’ to ‘the actual use of language in concrete situations’. Chomsky’s primary goal has always been to provide a description of competence and the innate mechanism underlying it. Competence is minimally defined in so far as the notion covers only that knowledge which every normal adult NATIVE SPEAKER of a language has of it. From this perspective, ‘The real difference between child L1 and adult L2 ACQUISITION is that in the former everybody ends up in the same place; in the latter…this is far from being the case’ (Schachter, 1996:86). Most of the extensive L2 learning literature which results from attempts to demonstrate and explain this difference agrees that L1 competence is achieved largely on the basis of innate properties of the language faculty. There is, however, considerable disagreement about the role of these properties in the learning of a second language after the so-called ‘critical period’ (see also AGE FACTORS).
The mature competence of a native speaker is the ‘steady state’ of their language faculty, reached, on the basis of its initial state which is genetically determined, after passing through a series of states in early childhood. The rate and route of this development is partially determined by the innate properties of the language faculty and partly by the typology of the language to which the person is exposed. The theory of the initial state is called ‘UNIVERSAL GRAMMAR’ (UG), and the theory of the steady state is called ‘GRAMMAR’ (Chomsky, 1995:14).
Each state is a state of knowledge, a mental phenomenon, unavailable for direct inspection. In formulating both UG and grammar, the linguist is therefore forced to rely on speaker-hearer performance. However, competence would only be directly reflected in performance in the case of ‘an ideal speaker-listener, in a completely homogeneous speech-community, who knows its language perfectly and is unaffected by such grammatically irrelevant conditions as memory limitations, distractions, shifts of attention and interest, and errors (random or characteristic) in applying his knowledge of the language in actual performance’ (Chomsky, 1965:3). In reality, what both a theorising linguist and a language acquiring child need to do is to extract from the varied examples of natural language use that they hear around them just those aspects that are relevant to the formulation of competence.
In this, the child is helped by the language faculty; UG determines the ‘principles’, or possible forms of human language, and also the ‘parameters’ within which it is possible for them to vary (Chomsky, 1981). In other words, UG restricts grammars powerfully, and ‘the theory of language acquisition will be concerned with acquisition of lexical items, fixing of parameters, and perhaps maturation of principles’ (Chomsky, 1995:28). The linguist may be helped by various forms of experimentation, particularly native speaker judgements of the grammaticality of sentences. From the point of view of the theory of second or foreign language (L2) teaching and learning, the interesting question raised by this view of first language (L1) acquisition is whether and to what extent ADULT LANGUAGE LEARNERS are able to access UG and use it in the development of a competence for L2. The answer might have implications for teaching methodology. UG might, for example, predispose a certain rate and route of learning of a second language, in which case it would make sense to structure teaching accordingly.
There are, as Mitchell and Myles (1998:61) say, four major views on this issue. According to the ‘No Access’ or ‘Fundamental Difference’ hypothesis (Clahsen and Muysken, 1986; Bley-Vroman, 1989), adult language learners have no access to UG. In favour of this hypothesis, Bley-Vroman (1989:43–9) points out that adult learners differ from child acquirers in failing to various degrees to become native-like; following different routes; using different strategies; having different goals; having unclear intuitions about grammaticality; benefiting considerably from instruction; and being susceptible to influence from affective factors. He argues that adults learn a foreign language, to the extent that they do, in reliance on the first language, which ‘fills the role which Universal Grammar has in child language acquisition’ (Bley-Vroman, 1989:42), and on their general problem solving abilities.
According to the ‘Strong Continuity’ (Epstein, Flynn and Martohardjono, 1998) or ‘Full Availability’ (Gair, 1998) hypotheses, in contrast, L2 learners have and use the same type of access to UG as L1 learners do (Krashen, 1981). Adherents claim that this hypothesis offers the best explanation for the ability of learners to reset parameters to suit an L2 which has different parameter settings than L1 (Al-Kasey and Pérez-Leroux, 1998).
According to the hypothesis of ‘Indirect Access’, L2 learners access UG via their L1 knowledge. There are two versions of this hypothesis (White, 1989:48). According to one, only those aspects of UG which are instantiated in L1 can be accessed, so parameter re-setting is impossible. According to the other, learners begin by assuming that L2 is like L1, but when they discover that this is wrong they can access UG to reset the parameters (White, 1986; Schwartz and Sprouse, 1994).
Finally, according to ‘The Weak Continuity’ (Vainikka and Young-Scholten, 1991, 1998), or ‘Partial Access’ hypothesis, only some aspects of UG are available to L2 learners. This position might help reconcile some of the conflicting data procured by studies seeking confirmation or falsification of the remaining positions. However, as Mitchell and Myles remark (1998:68), these data themselves tend to form the subject matter of the continuing debate concerning access to UG in L2 learning (see, e.g., Schachter, 1996), and the jury remains out on the question of the exact role of UG in second language learning.