In linguistics, linguistic competence is the system of unconscious knowledge that one knows when they know a language. It is distinguished from linguistic performance, which includes all other factors that allow one to use one's language in practice.
In approaches to linguistics which adopt this distinction, competence would normally be considered responsible for the fact that "I like ice cream" is a possible sentence of English, the particular proposition that it denotes, and the particular sequence of phones that it consists of. Performance, on the other hand, would be responsible for the real-time processing required to produce or comprehend it, for the particular role it plays in a discourse, and for the particular sound wave one might produce while uttering it.
The distinction is widely adopted in formal linguistics, where competence and performance are typically studied independently. However, it is not used in other approaches including functional linguistics and cognitive linguistics, and it has been criticized in particular for turning performance into a wastebasket for hard-to-handle phenomena.
Competence versus performance
Linguistic theory is concerned primarily with an ideal speaker-listener, in a completely homogeneous speech-community, who knows its (the speech community's) 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 this language in actual performance. ~Chomsky,1965(p. 3)
Chomsky differentiates competence, which is an idealized capacity, from performance being the production of actual utterances. According to him, competence is the ideal speaker-hearer's knowledge of his or her language and it is the 'mental reality' which is responsible for all those aspects of language use which can be characterized as 'linguistic'. Chomsky argues that only under an idealized situation whereby the speaker-hearer is unaffected by grammatically irrelevant conditions such as memory limitations and distractions will performance be a direct reflection of competence. A sample of natural speech consisting of numerous false starts and other deviations will not provide such data. Therefore, he claims that a fundamental distinction has to be made between the competence and performance.
Chomsky dismissed criticisms of delimiting the study of performance in favor of the study of underlying competence, as unwarranted and completely misdirected. He claims that the descriptivist limitation-in-principle to classifying and organizing data, the practice of "extracting patterns" from a corpus of observed speech, and the describing of "speech habits" are core factors precluding the development of a theory of actual performance.
Other generativists
Linguistic competence is treated as a more comprehensive term for lexicalists, such as Jackendoff and Pustejovsky, within the generative school of thought. They assume a modular lexicon, a set of lexical entries containing semantic, syntactic and phonological information deemed necessary to parse a sentence. In the generative lexicalist view this information is intimately tied up with linguistic competence. Nevertheless, their models are still in line with the mainstream generative research in adhering to strong innateness, modularity and autonomy of syntax.
Ray S. Jackendoff
Ray S. Jackendoff's model deviates from the traditional generative grammar in that it does not treat syntax as the main generative component from which meaning and phonology is developed unlike Chomsky. According to him, a generative grammar consists of five major components: the lexicon, the base component, the transformational component, the phonological component and the semantic component. Against the syntax-centered view of generative grammar(syntactocentrism), he specifically treats phonology, syntax and semantics as three parallel generative processes, coordinated through interface processes. He further subdivides each of those three processes into various "tiers", themselves coordinated by interfaces. Yet, he clarifies that those interfaces are not sensitive to every aspect of the processes they coordinate. For instance, phonology is affected by some aspects of syntax, but not vice versa.