英语论文网

sity College, London, and at the time of writing professor at the University of Sydney, and R. A. Hudson (b. 1939), of UCL. Syntactic analysis in the London style is commonly called “systemic grammar” (other, less significant terms have also been used). A “system” in Firthian language, remembers, is a set of mutually exclusive options that come into play at some point in a linguistic structure. This is the clue to London School syntax: like Firthan phonology, it is primarily concerned with the nature and import of the various choices which one makes (consciously or unconsciously) in deciding to utter one particular sentence out of the infinitely numerous sentences that one’s language makes available. To make this clearer, we may contrast the systemic approach with Chomsky’s approach to grammar. A chomskyan grammar defines the class of well-formed sentences in a language by providing a set of rules for rewriting symbols as other symbols, such that if one begins with the specified initial symbol S and applies the rules repeatedly the end-result will be one of the target sentences. Such a grammar can succeed in defining a range of different sentences, clearly, only because in applying the rules one is often faced with choices. But in a Chomsky grammar the choice-points are diffused throughout the description, and no special attention is drawn to them. Many choices are made in the constituency base: a given category symbol is expanded by means of braces or commas into alternative rewriters, or brackets are used to show that some element may or may not occur in the rewriter of a category symbol. Other choices arise in applying transformations: certain transformations are optional, others can apply in alternative ways, and (in some versions of transformational theory) there are alternative orders for applying transformations, with the nature of the ultimate result varying according to which order is selected. Often it would be the case that some choice in applying transformational rules becomes available only if certain options have been selected in the constituency base, but a Chomskyan grammar does nothing to make such interdependencies between choices explicit – that is not its aim. To cite a very simple example, Halliday (1967, p. 40) suggest that one system of choices operating in English main clauses, a system which he labels “transitivity”, provides for a choice between “intensive” and “extensive”. In a standard transformational grammar, the syntactic differences between these clauses would correlate with choice of rewrite for the category symbol “VP” and for certain other symbols in the base, with choice of whether or not to apply the Passive transformation, and with choice of whether or not to apply the transformation which deletes the by-phrase produced by passive. No explicit statement would be found in a transformational grammar pointing out, for example, that the choice of applying the Passive transformation arises only if certain options are chosen when rewriting “VP” in the base, and there are certainly no special names given to the alternative structures which result from the various choices. (Occasionally Chomskyans do use a special terms to describe some particular syntactic structure, but usually this is a term inherited from traditional philological vocabulary, and traditional terminology provide names for only the most elementary among the many systems defined in a systemic grammar – Chomskyans do not make a point