REVIEWS
ROSAMUND MOON, Fixed expressions and idioms in English: A corpusbased approach. (Oxford Studies in Lexicography and Lexicology.) Oxford:Clarendon Press, 1998. xiii + 338 pp.
Reviewed by ELEANOR 0LDS BA TCHELDERThis is a book for all data-oriented linguists. Moon offers a wealth of information about fixed expressions and idioms (FEis) in English. The book reports the results of a detailed study of the "frequencies, forms, and functions" (p. 1) of FEis found in a single large corpus, "incorporating distributional, formal, semantic, and discoursal information" (p. 56).The book is based on Moon's doctoral thesis (University of Birmingham, 1994), for which she created a database of 6776 FEis. Most of these were expressions which had been identified for the Collins cobuild English language dictionary ( CCELD ), of which she was an editor, and then others were added from various sources as the study proceeded. We might ask whether this is a large or small sample:I intended to include, as far as possible, a large proportion of the commonest FEis in current British English, together with some commoner FEis from American English. The database does not of course record the complete set of the FEis of English, which is uncharted, unquantified, and indeterminate (p. 44).It is difficult to compare the numbers with other listings, since the criteria for inclusion often differ. In any case, since Moon's goal here is to describe the phenomena largely from a qualitative rather than a quantita- 263
264WORD, VOLUME 52, NUMBER 2 (AUGUST, 2001) tive perspective, the fact that her data may be merely representative rather than exhaustive is not a serious drawback.The items studied were a variety of multi-word lexical items, excluding hyphenated words, compounds, phrasal verbs, foreign phrases, and multi-word verb inflections (had been given). Also excluded were collocating words that do not form a syntactic unit (of the). Three principal criteria for the "holism of a string" (p. 6) were used: institutionalization (lexicalization), assessed by the frequency with which the string recurs in the corpus; lexicogrammatical fixedness, or rigidity of form; and non-compositional meaning or semantic idiosyncrasy. Many of these terms and concepts are problematic, as Moon discusses at some length in Chapter 1.Despite the fact that "there is no generally agreed set of categories, as well as no generally agreed set of terms" (p. 19), Moon attempts a typology of FEis. Based on "whether the string is problematic and anomalous on grounds of lexicogrammar, pragmatics, or semantics" (p. 19), she defines three main category groups (p. 19ff. and For me, these categories were rather unsatisfying. As I read through the book, they never became cohesive or intuitive, but to the end seemed rather arbitrary. Moon herself does not make any great claims for them: "This typology is simply a means to an end: a way of classifying a wide range of FEis so that ... global statements could be made about broad groupings of FEis" (p. 20).Chapters 2 and 3 d...
