[an error occurred while processing this directive]
Commentary Logo
Japan Island disputes in Southeast Asia    World War III Alarm Anti-Islam video and anti-US sentiment    Skull Say no to brainwashing

       peace and anti-war Israel Iran strike imminent    sanctions as a result of nuclear tests Sanctions on Iran and workarounds    black lightning bolt India's massive blackout

Snapshots of news
Gearheads and mastheads
SHORTCUT TO MAIN SECTIONS & ARTICLES
Home  •  About this site  •  How did we once fight corruption in colonial Hong Kong?
 •  Historical US administrative thoughts  •  USA versus colonial Hong Kong
 •  Anti-corruption review of Nigeria  •  Procurement monitoring in Nigeria
 •  Syndicated news  •  Usman's blog  •  Anti-graft news  •  Socialist news

[an error occurred while processing this directive]
[an error occurred while processing this directive] [an error occurred while processing this directive]

John Sinclair's lexical items – an introduction

Raymond Cheng

Add Text Here 2 (c) Floyd Anderson
Photo © Floyd Anderson

In order for us to be able to move onto our analysis with the historical news corpus (in later sections), we will first need to examine the origins of the five categories of co-selection that make up the late Birmingham University professor John Sinclair's "lexical term" (Sinclair, 2004).

Open-choice model and the idiom principle

The five categories of co-selection of a lexical term originates from Sinclair's idiom principle (1991) in which he emphasized the "non-random nature of language" as oppose to his open-choice model (also known as "slot-and filler") where he sees language as a product "of many complex choices". Sinclair described his open-choice model as "probably the normal way of seeing and describing language" (1991, p. 109). This model, in general, describes the basic restrictions on the possible choice of lexical items that can be used to syntactically fill in every "next vacancy" appearing in any given text. The model, hence, forms the basis of most uses of the grammar. On the contrary, Sinclair's idiom principle is based on the presumption that users of a language select naturally from a set of pre-constructed phrases, or idioms, as the name of the model implies. With the two theories presumably going in opposite directions, no one theory seems to be able to provide one comprehensive explanation. Statistically speaking, Erman and Warren (2000) found in their research that about 45% of all authentic text goes with the open-choice model, whereas the remaining 55% (a little more than half) belongs to the idiom principle. But this does not mean that these two models are competing against with one another. In fact, according to Sinclair's original concept, the two theories actually work hand-in-hand in which case the open-choice analysis "could be imagined as an analytical process which goes on in principle all the time, but whose results are only intermittently called for." (1991, p. 114)

In 1996 and 1998, Sinclair further expounded his idiom principle to include the five categories of co-selection that make up the lexical item, namely, the invariable core word (or core words), the obligatory semantic prosody, the optional collocation and colligation from Firth (1957), and, last but not the least, the semantic preference.

[an error occurred while processing this directive]