We round off this brief visit to the dictionary by pointing out that we began with the stereotypical view that there are rules that specify syntactic constructions and that words are listed in the dictionary or lexicon, taken out of the dictionary and inserted into syntactic construc- tions. It turned out that the connection between syntax and lexical items is closer than we might have imagined, not only with respect to the number of connections but also with respect to the powerful role played by individual lexical items, especially verbs. Over the past fifteen years or so, analysts have come to realise that in any language there is a large set of phrases and even whole clauses that are not freely built up but fixed. Following the discussion by Crystal in The Cambridge Encyclopaedia of the English Language (pp. 162–3), we can distinguish various types of fixed phrase, as set out in (18)–(21).
- POLYWORDS: in a nutshell; once and for all
- INSTITUTIONALISED EXPRESSIONS: Give me a break; How do you do?; Long time no see; Pleased to meet you; See you soon; come to think of it; it doesn’t bear thinking about
- HIGHLY CONSTRAINED PHRASES: as I was (saying/
mentioning); as far as I (can see/can make out/know); I (just) can’t think straight; I thought I told you not to do that
- SENTENCE BUILDERS: my point is that ; let me begin by
; not only … but also
Crystal also discusses collocations,
restricted sequences of words; examples are in (22). Heavy smoker
and
heavy drinker qualify as colloca- tions because heavy and light combine with a limited set of nouns
– smoker, drinker, eater and sleeper.
THE LEXICON 57
- COLLOCATIONS: staple food/diet/industry; wouldn’t/won’t/ couldn’t budge; a heavy smoker; a heavy drinker
The above five types of phrase and clause do not always conform to the syntax of written English and may have idiosyncratic meanings; a heavy smoker is not a smoker with a weight problem. It looks as though these phrases and clauses must be listed as single items in the dictionary, although at the time of writing there are no explicit descriptions of English or other languages that handle fixed phrases in an adequate way. What analysts are more and more certain of is that the number of fixed expressions in English (and other languages) and the frequency with which they are used by speakers and writers is much greater than anyone supposed forty years ago.
