The manner in which a child acquires language is a matter long debated by linguists and child psychologists alike. The "father" of most nativist theories of language acquisition is Noam Chomsky, who brought greater attention to the innate capacity of children for learning language, which had widely been considered a purely cultural phenomenon based on imitation. Nativist linguistic theories hold that children learn through their natural ability to organize the laws of language, but cannot fully utilize this talent without the presence of other humans. Chomsky claims that children are born with a hard-wired language acquisition device (LAD) in their brains. They are born with the major principles of language in place, and with some parameters to set (such as whether sentences in the language they are to acquire must have explicit subjects). According to nativist theory, when the young child is exposed to a language her LAD makes it possible for her to set the parameters and deduce the grammatical principles, because the principles are innate.
This is still a controversial view, and many linguists and psychologists do not believe language is as innate as Chomsky argues. There are important arguments for Chomsky's view of development, however. These include the idea of universal grammar, the similarities that underlie every human language. Another argument is that without a propensity for language, human infants would be unable to learn such complete speech patterns in a natural human environment where complete sentences are the exception. This is known as the poverty of stimulus argument.
· explains why language is learned relatively quickly
· explains how language is learnt despite poverty of the stimulus
· very little evidence for adultlike grammatical knowledge in young children
· young children make errors Chomsky would not predict
· cannot explain why children make grammatical errors (e.g. doggie go walkies) even after extensive language exposure
Chomsky set out his theory of transformational grammar in Syntactic Structures (1957), a book that revolutionized the development of theoretical linguistics. In this work he broke with the dominant structural school, which held that language is essentially a system of syntactical and grammatical habits established by means of training and experience. Chomsky, by contrast, argued that human beings have an innate facility for understanding the formal principles underlying the grammatical structures of language. It is this innate capacity that explains how young children, after hearing the speech of their elders, are able to infer the grammatical rules underlying ordinary sentences and then use those rules to generate an infinite number and variety of sentences that they had never heard before.
In analyzing the innate ability to construct these "generative grammars," Chomsky distinguished between two levels of structure in sentences: "surface structures," which are the actual words and sounds used, and "deep structures," which carry a sentence's underlying meaning. People are able to create and interpret sentences by generating the words of surface structures from deep structures according to a set of abstract rules that, though limited in number, allow for unlimited variation. Chomsky called these rules "grammatical transformations," or "transformational rules." He argued that these rules are basically the same in all languages and correspond to innate, genetically transmitted
Chomsky was able to show that both phrase structure grammar and transformational grammar are more powerful (i.e. can do more) than finite state grammar, and that transformational grammar is a more powerful grammar than phrase structure grammar. Transformational grammar is essentially Chomsky's own contribution to a general theory of grammar. The other two grammars - although previously not formalised - existed in linguistics prior to Chomsky's work. Only a transformational grammar can derive the basic rules constitutive of the ideal speaker-hearer of, for example, English. The logic behind transformational grammar is that if every utterance implied a unique rule as a condition of its acceptability, there would be too many rules to deal with. Clearly, the number of rules are not equivalent to the number of utterances; this is what any grammar implies. On the other hand, Chomsky points out that if one cannot show that many sentences - apparently different at a 'surface' level of phrase structure grammar - are in fact transformations of the same rule, the grammar becomes almost infinitely complex and contains little explanatory power. Phrase structure grammar would thus become too complex if it alone were charged with providing all the rules of the ideal speaker-hearer's sentence formation. In sum, then, a transformational grammar is a way of reducing sentence formation to the smallest number of rules possible. From a slightly different angle, the transformational grammar, providing the rules of competence, is equivalent to Chomsky's notion of 'deep structure'.
Language learning is viewed as hypothesis formation and rule acquisition, rather than habit formation. Grammar is considered important, and rules are presented either deductively or inductively depending on the preferences of the learners. Class exercises are intended to give learners ample practice with rule application. Errors are viewed as the inevitable by-products of language learning. Error analysis and correction can be seen as appropriate classroom activities from which both teachers and students can learn. The focus is still largely sentence-oriented.