Retorts

Fat Idioms

I read Koenraad Kuiper's article on idioms, in your November issue, with interest. If it had been written by a "hard" scientist, it probably would have featured some analysis of the idiom database of 20 000 entries, which he refers to, but doesn't seem to have interrogated. Some sort of frequency distribution would have shown whether our idioms are in fact "fat" or "thin", say, without postulating arbitrary upper limits on numbers of lexical items and so on. To hypothesise in advance of the data, as Sher lock Holmes said, is to find oneself up shit creek in a barbed-wire canoe

Ken Harrop, Auckland