<\/p>\nDictionary of Selected Collocations<\/h2>\n
Jimmie Hill and Michael Lewis, Eds. (1997)
\nHove, UK: Language Teaching Publications
\nPp. 288
\nISBN 1-899396-55-1
\nUK \u04088.95 <\/b><\/p>\n
Assuming that a collocation is at the very least a pair of words which occur together with a significant degree of probability, curiosity would ask of any collocation dictionary how such pairs could be organized so as to avoid double entries. The organizational principles of Dictionary of Selected Collocations (DOSC),<\/i> then, are perhaps the first point of interest for a reviewer.<\/p>\n
Essentially, the layout reflects the editors’ recommendation that users of this dictionary be well past the intermediate stages of language learning. As well as alphabetically, headwords and listings are organized and then sub-divided by word class. Depending on whether the search for a collocation starts from a noun or from a verb or adjective, the learner looks up a headword in either the Noun or the Adverb section. These two sections, aside from the tribute, introduction, and a two-page list of sentence adverbs, comprise the bulk of the dictionary.<\/p>\n
The learner will need some degree of prior semantic knowledge about both halves of the collocation. No glosses are provided, either for headwords or entries, except for those very few instances when a headword has been editorially judged as having “clearly distinct” meanings; e.g., talk<\/i> (lecture), talks<\/i> (negotiations). (It would be interesting to test-run the dictionary with learners, especially to determine the practicability of DOSC’s<\/i> contention that, despite the absence of glosses, learners can nonetheless select collocations through a culling process whereby familiar words are contemplated as potential candidates and unknown words ignored. Unfortunately, the present reviewer’s young EFL learners would most likely be scared off by any kind of book that, upon first glance, showed more words than pictures.)<\/p>\n
The layout aside, a second point of interest here are DOSC’s<\/i> opening pages. Someone not aware of the revolution that has taken place in modern lexicography may wonder why DOSC,<\/i> given the usefulness of collocational knowledge, has taken up precious lexicographical space for self-justification and self-approbation. On the other hand, those who are in the know, not only about the changes in lexicography but about the increasing significance of lexis to language education and analysis, may recognize in DOSC’s<\/i> introduction and guide a manner and tone typical of most of the other dictionaries recently published under the banner of being non-traditional. Some critics might go a step the other way and [-1-] accuse DOSC<\/i> of not going far enough in making its selection process transparent.<\/p>\n
On its cover, DOSC<\/i> states that it offers 55,000 collocations under 3,200 essential headwords to help learners make more natural and hence better use of words already somewhat known. As qualified in its title, however, DOSC<\/i> is made up of a list of selected collocations, for even when narrowly defined as a two-constituent unit, the total number of English collocations exceeds many times over the number DOSC<\/i> lists. Excluded are those collocations which, the editors write, are too common, too technical, too colloquial, or too difficult for learners to use; included are those which are in a strong relation, the judgment of strength presumably resting upon the editors’ expertise and their native speaker inclinations.<\/p>\n
The judgment call is not theirs alone, however. It is a well-known custom in lexicography for a dictionary’s current editors to pay tribute to the continuous and accumulated efforts of their predecessors, and though DOSC<\/i> is a first edition dictionary, its editors maintain tradition by acknowledging the two Warsaw academics from whose work and publications in the 1980s DOSC<\/i> is derived: C. D. Kozlowska and H. Dzierzanowska. It is their thoughts and impressions too which helped inform the selection process and which DOSC’s<\/i> editors humbly acknowledge as ahead of the times.<\/p>\n
Others may indeed have been slow in coming forth on paper with their collocational opuses (on-line is another matter, but that is another review). The source of the tardiness, however, was caution rather than neglect, a caution which suggests in return that DOSC’s<\/i> publishers might have been a tad more foolhardy than brave for having crossed the posts sooner than safer.<\/p>\n
Michael Lewis and Jimmie Hill, editors and founders of LTP, have evinced a long-time interest in and a well-known defence of the importance of lexis to language education and analysis, and are not unlikely therefore to be unaware of how researchers and theorists, even before the 1980s, estimate collocational validity. As editors, though, Lewis and Hill may have had a more immediate concern with elucidating for learners the whys and hows of collocations in simple and terminology-free English, rather than with pleasing in an empirical manner those who are corpus-linguistically in the know. Admittedly, when leafing at random through DOSC,<\/i> its collocations do seem to ring a bell (who knows, however, what internal statistical skewing has been effected by ten years spent in non-English households reading mostly pre-20th century writers), and so might very well be a boon for advanced English learners looking for ways of making their English more natural. Ultimately, however, significant details about the corpus and the selection process underlying both the keywords and the collocations ought to be made available for those holding qualified doubts about whether DOSC<\/i> indeed typifies what it claims is natural English, always such a [-2-] protean thing. After all, the line of fascinating work that saw birth decades back when J. R. Firth and his chain of disciples first dreamed up the term collocations has only now begun, by virtue of computer-driven third-generation corpora which run into the hundreds of millions of words, to confirm in statistically defensible ways our linguistic impressions, collocational or otherwise.<\/p>\n