Review of Wells Longman Pronunciation Dictionary. Third edition. 

This version incorporates some very slight changes from the original one which appeared in 2009 in Vol 39 No 2 of the  Journal of the International Phonetic Association  published and copyright by Cambridge University Press

1. This dictionary, in brief LPD, appeared originally in 1990 and was first revised in 2000. This second revision involves no fundamental changes though it incorporates 3,000 new headwords and a variety of modest improvements. This review will compare it briefly with the two other equivalent works available: EPD, the Cambridge English Pronouncing Dictionary and ODP, the Oxford Dictionary of Pronunciation for Current English.

2. The LPD extra headwords have not increased its bulk. In dimensions EPD and LPD are the same except that LPD is slightly (1 cm) shorter. ODP is like LPD except for being (1.5 cm) thicker. LPD newly matches EPD's provision of a CD-ROM "with all words and phrases spoken aloud in both British and American English". It also contains some instructional materials. ODP has no disc.

3. LPD is probably an optimum size. A larger work would only weakly compete with large general dictionaries such as the OED (Oxford English Dictionary). That work should better have used Sweet's Romic pronunciation notation, the basis of this Association's alphabet, but it adopted a system of excessive complexity which remained until 1989 when its second edition transferred to the IPA alphabet. The OED third edition, which can be seen online as so far drafted, is now at last receiving attention to pronunciation matters of reassuring quality and fullness.

4. LPD, EPD and ODP all provide both British and American coverage. None of them offers anything much on Scottish, Irish, or Commonwealth usages though LPD, unlike EPD, records "pronunciations widespread in England among educated speakers" but "judged to fall outside RP". ODP provides unacceptably subjectively selected mixtures as its own models.

5. Among the changes to LPD are some to words with the prefixes be-, de-, e-, re- and se- so that, when unstressed, these are now usually shown with the cover symbol i standing for "pronounced indifferently with /ɪ/ or /iː/" (p. xiii). These are rightly not completely blanket changes as can be seen by comparing eg event with select and recast with revise. The choice of [i] rather than [ɪ] or [ə] in respect of some common words, eg believe and remind, may not meet with universal assent. Two other changes, mentioned at p. xiii, are the acceptance of /tʃuː/ and /dʒuː/ as alternants of /tjuː/ and /djuː/ and the representation of American <wh-> words as having sub-variant rather than co-variant /hw-/ alternants.

6. LPD's Introduction (p. xvii) makes it clear that it is offered principally as a reference book for teachers and learners of English: it gives at each entry a "main" pronunciation which is that form "recommended for EFL purposes". EPD says "The first pronunciation given is believed to be the most usual one..." (p.vi). ODP is not explicit about this but declares (p.viii) that it "is intended for use ... by learners of the language".

7. Always simply indicating what pronunciation is judged to be predominant in native-speaker usage may well not be the most appropriate course to follow for works with a positive EFL orientation. There can surely be little argument that, if there are two fully common variants in existence, recommending the one which is not antagonistic to the spelling is beneficial to the learner. One imagines that this principal has operated in at least some of the following examples. All three works show first the version of government with an /n/ which native speakers most often omit. They all show I'll with the /aɪ/ diphthong though it is probably more often heard as /ɑːl/. They all completely omit the very common forms of infrastructure, prescription and prerogative which contain only the second /r/ of their spelling.ᵻ Other largely predominant British usages that mostly do not receive even acknowledgement of their existence include accept with /ɪk-/, hospital with/-dl/, obviously, previously, seriously etc with no yod, certainly, suddenly and only with no /l/, and plurals ending -sts with no final /s/. LPD gives private with /-ət/ first though three-quarters or so of speakers use /-ɪt/: a rule of thumb for learners to use schwa for all -ate adjectives could be said to justify that ordering; EPD gives /-ɪt/ priority in line with its stated policy.

8. Apart from its word list, LPD contains what amounts to a condensed textbook on English phonetics in the form of about forty information "panels" incorporated in the alphabetical sequence at their titles. There are also numerous panels on sound-to-spelling correspondences beginning the sections dealing with each letter. Many prefixes and suffixes are entered with notes on their effects on stress patterning: for example there are twenty lines on trans- and a dozen on -ment. EPD has similar features but they are much less extensive; ODP has nothing similar. LPD has 260-odd graphics showing relative scores for different pronunciations in opinion polls of reader preferences. These newly incorporate effective pie charts. LPD is better provided with vowel diagrams than EPD which has no American ones ; ODP has none at all. LPD and EPD accounts of the weakforms of function words, valuable for the advanced learner, are excellent. ODP is weak on them.

9. Inevitably LPD contains features that will not be to everyone's taste. As to syllable-division indications, the LPD spaces may seem legibility-impairing or counter-intuitive to some just as for others the IPA dots in EPD may seem to produce more clutter than they are worth; like the EPD of Jones's day, ODP has none. Some may think that various features could be dealt with better by simply giving rules eg the subscript voicing symbol ˬ for American tapped /t/ (ODP shows this as /d/), the wedge symbol for stress shift, the arrow → for automatic variants. Some may feel that the special uses of exclamation marks ("unexpected for this spelling"), asterisks (flagging up important American differences) and equals signs (identifying homophones) are plethoric. Others may feel there are too many disapproved forms: there are more than in EPD and ODP. LPD admits "intrusive" r's without comment as does ODP; EPD simply omits them eg at withdrawal. Others again may object to the inclusion of [ɒʊ] as too markedly southeastern. Some may blench at the sight of such complex pieces of notation as that for instance of the word temporarily [ˈtemp ərər əl i] with its three spaces to indicate syllable divisions, three schwas two of which are italicized and one in raised small type, plus a potential-compression loop. Compare EPD's temporary [ˈtem.pər.ər|.i] with its three syllable-division dots, raised small schwas and vertical-line cut-back sign (the adverb not being given a separate entry). ODP's temporarily [ˈtɛmp(ə)r(ə)rɪli] has two bracketed schwas and a non-IPA symbol (barred ᵻ), after Kenyon & Knott, covering the alternatives /ɪ/ and /ə/, a neat device but less than perfectly legible as printed.

10. General legibility is good in all three even though LPD has two, EPD three and ODP four columns per page. LPD has been newly upgraded for legibility. The introductory matter is set in a distinctly larger typeface than previously and the headwords preceding the phonetics are now set (still bold) in a fairly bright blue which proves to be a more effective use of colour than in the second edition where the non-bold phonetics coming after the headwords tended to look comparatively faint. Now they are clearer partly because the "main" pronunciations are in bold black. This is the same style as had been already used effectively in EPD but LPD's new larger type makes it very clear. Unlike EPD, LPD now uses bold colour for its selections of compounds and phrases which both books provide in paragraphs after certain headwords as a space-saving expedient. ODP certainly has greatest overall legibility but its separate lines for variants are achieved at the cost of less full coverage than the other two and greater bulk. It has the added disadvantage of several vowel symbols pointlessly failing in compatibility with general practice and dismayingly irrational correlative representations of the price and mouth diphthongs. See Windsor Lewis 2003 (§5.1 on this website).

11. This LPD third edition consolidates one's impression that, even though EPD is a strong rival and ODP is undoubtedly easier to read, this is simply the best ever dictionary of English pronunciation. A feature of special note to IPA readership is LPD's provision at numerous loanwords of excellent transcriptions of their pronunciations in over fifty foreign languages.

References

Jones, Daniel. 2006. Cambridge English Pronouncing Dictionary edited by Peter Roach, James Hartman & Jane Setter ("EPD").

Kenyon, John S. & Thomas A. Knott. 1944. A Pronouncing Dictionary of American English. Springfield, Mass., USA: Merriam Co.

Murray, Sir James A. H. & al. (1933;1989). The Oxford English Dictionary. Oxford: Clarendon Press ("OED").

Upton, Clive, Wm A. Kretzschmar Jr & Rafal Konopka. 2001. The Oxford Dictionary of Pronunciation for Current English. Oxford: OUP ("ODP").

Windsor Lewis, Jack. 2003. IPA vowel symbols for British English in dictionaries. Journal of the International Phonetic Association 33(2), 143-152.