UNIVERSAL GRAMMAR (UG)
The term is used in two ways:
a. A mechanism which is innate in human beings and which sensitises an infant to the features which characterise all languages, thus giving it a head start in the process of acquiring its own. Some accounts see UG as part of a maturational process, with awareness of the characteristics of language gradually unfolding as the child develops. Others assume that it is fully present, hard wired, from the start, but that the child is not able to take full advantage of it because of its limited cognitive development.
b. The linguistic content of such a mechanism, a set of phonological, syntactic and lexical features which are shared by all languages. The content of UG falls into two categories: principles which are true universals, occurring in all (or nearly all) languages, and parameters, which are universal to the extent that they occur across languages in one form or another. An example of the latter is the pro-drop parameter which specifies whether a language employs an obligatory subject pronoun (She speaks English) or is able to omit it (Parla inglese). Another example is seen in the fact that languages with a standard VO (verb + direct object) word order make use of certain syntactic patterns which differentiate them sharply from those with an OV (direct object + verb) order. These patterns constitute a set of ‘If. .. then’ features– meaning that if a language has one, it is likely to manifest most or all of the others.
There are two broad approaches to the quest for evidence of universals. The Chomskyan approach is based upon putting to the test a detailed theoretical model. Researchers might ask adult subjects to make grammaticality judgements, deciding whether a particular string of words is or is not acceptable. These judgements are said to tap in to competence and are preferred to evidence of actual speech production, which would reflect performance. An alternative typological approach examines evidence across many languages in order to trace similarities. There is a broad measure of agreement that the following are universal:
some lexical categories (noun and verb);
structure-dependency;
phrases containing a head of the same type as the phrase;
a phrase structure consisting of Specifier, Head and Complement.
UG theory accepts that languages may deviate to some degree from the universal pattern. A language user’s competence is said to consist of a core grammar of universal principles and parameters and a periphery of features specific to the language in question, which cannot be explained by reference to UG. They might be survivals from an earlier stage of the language, loans from other languages or fixed idioms. The relationship between core grammar and periphery is best described as a cline, with central UG characteristics defined as unmarked and features that fall outside UG as progressively more and more marked. When acquiring a language, the child comes to recognise that some features of the target language do not conform to the criteria specified by UG, and have to be mastered by a process which is independent of the normal acquisition route.
Chomsky himself recognises that cognitive constraints such as short-term memory restrictions may have an impact upon performance, limiting what the child is capable of achieving with its innately acquired UG. For example, the absence of a subject pronoun in much early English speech might indicate a parameter setting which initially favours a pro-drop pattern; but it might equally well indicate a cognitive inability to process more than two words. A useful distinction has been made between language acquisition, which is supported by UG, and language development, in which cognitive factors play a part.
See also: Chomskyan theory, Language acquisition: theories, Language universals, Markedness, Principles and parameters
Further reading: Cook and Newson (1996)
