Tuesday 18 April 2017

100 years of the English Pronouncing Dictionary

As we approach English Language Day on 23rd April, I thought it would be a nice idea to write a short blog post about the English Pronouncing Dictionary (EPD), which I co-edit with Peter Roach (principal editor) and John Esling (American English, from the 18th edition). This is especially salient as it is celebrating its 100th birthday this year, being first published in 1917. We marked this at a special Pre-Conference Event of the IATEFL Pronunciation Special Interest Group and a Cambridge University Press event at the 2017 IATEFL Conference in Glasgow.

The EPD was created by British phonetician Daniel Jones, who was head of the Department of Phonetics at University College London. Jones is credited with coining the term 'phoneme' in 1917, too, so it was a bit of a special year all round for the subject area. Jones had collaborated on a dictionary project prior to the EPD but, rather than listing headwords orthographically in alphabetical order, that version had listed the headwords in phonemic script first, with the spelling form following. It was not a best-seller.

Daniel Jones.
Image: https://evolution9linguistics.wordpress.com/2010/01/03/dj-system-kk-system/
The EPD was first published by Dent, who continued to produce it until the late 1980s, when it was bought by Cambridge University Press (CUP). During the Dent years, Jones produced a further 11 editions, with A. C. Gimson stepping in as editor following Jones's death in 1967. Gimson produced two editions, the 13th as sole editor, and the 14th with the support of Susan Ramsaran, who finished the work following Gimson's death in 1985.

I first got involved with EPD in its 15th edition. I'd been doing an MA at Leeds, where Peter was based at the time, and was invited to join the team; I am listed as 'Pronunciation Associate' on the title page of EPD15. From this edition, it was decided to add American English pronunciation as well as British English, and so James Hartman was brought on board. The other exciting thing about this edition was that it was being computerized using the impressively-named 'Advanced Revelation' database software. My main work at the time was to go through all the existing pronunciations ('prons') in the database to check they were up to date and entered properly, syllabify them according to the principle of Maximal Onsets, and add new prons for words coming in from CUP after consultation with Peter and Jim; we added more than 18,000 new words at the time. What that usually entailed was me researching and suggesting both British and American prons and Peter and Jim agreeing, disagreeing or augmenting the suggestions with additional variants. 

Apologies for the huge understatement, but the English language has changed rather a lot since 1917, pronunciation included. Peter was very much against using the term RP to describe the reference accent in EPD (now CEPD) as he perceived it to be outdated and associated with the upper classes. For British English, he prefers the term 'BBC English', and for American English we use 'Network English'. The idea is that these are the accents used by professional speakers on national broadcast networks in the UK and USA; these are educated speakers who could come from any demographic.

Cambridge University Press have produced an engaging video on CEPD to mark its 100th year. I love the way the narrator trills his /r/ when he pronounces 'American'!

So, what next for CEPD?

There are no current plans to produce any further print editions of the dictionary; it is now available through the Apple and Android app stores (accessible via the video above), and CUP have told us that the way forward (as I write) is electronic editions only. However, as well as adding new words to the dictionary from time to time, there may come a point at which we will have to evaluate whether BBC or Network English are relevant anymore. John Wells, Emeritus Professor of Phonetics at UCL, has written of EPD that it 'has set the standard against which other dictionaries must inevitably be judged'. In order for that to continue, we may want to add prons for something like Standard Global English - whatever that might look like - or make changes to the current transcription systems to reflect the pronunciation of future Englishes. The phonemic system of transcription is fairly robust and forgiving, but it wasn't so long ago that we added happY and thank yOU vowels, which are non-phonemic. 

Will we want to add glottal stops, for example? 

And can the CEPD continue to be a suitable reference point when English is developing so fast around the world? 

I'd be interested to hear your thoughts.


  1. I think the way forward is to recognize that there is no standard pronunciation any more (most notably in the U.S., but elsewhere too), and to use cross-accentual prons (nice word!) based on an expanded version of John Wells's lexical sets, so that the speakers of Standard English in every accent can figure out how to pronounce words in their own accent rather than the way some other speaker would pronounce them.

    In principle, every word in the vast Standard English vocabulary should be accessible to everyone who speaks it, whether their accent is RP, Southern American, Ottawa Valley, New Zealand, Broad Australian, or Northern Indian. (There would be exceptions for words whose pronunciation is unique, like the General Australian gone /gɔːn/.

    Yes, it would be a lot of work, but for a huge amount of return.

    1. I agree that every word should be accessible to everyone who speaks it. I'm not sure a pronouncing dictionary can represent multiple accents, though. Your suggestion about cross-accentual prons sounds good, but I wonder how one might action it.

  2. Hi, Jane
    I have been using the English Pronouncing Dictionary for quite a long time now and I don’t think I will ever get used to those electronic editions –I love the touch and smell of paper.

    1. Yes, I prefer them, too! Peter and I are lobbying for further print editions but these things are always down to cost, sadly.

  3. Dear Ms. Setter,
    Thank you for your article on the centennial of the CEPD. I still have the first copy I bought back in the 90s. It actually came as sort of a shock. I had been studying English (in Italy) for a long time and I thought I had the hang of it. How mistaken I was. When I started checking the pronunciation of some very common everyday words I thought there was something wrong with that dictionary, that it was full of typos or something. I had been mispronouncing so many words that it couldn't be right. Alas it was. An eye opener. And it still is every time I use it in class with my adult learners.
    I would hate it if further editions didn't see the light of day.
    I think the best combo is the paper edition with cd-rom (it is so useful to have it installed in your pc) with perhaps an online extension. I haven't tried the App myself.
    As for the kind of pronunciation it should reflect I think that BBc English and Network English are, for the time being, still the way to go. I mean there must be a starting point somewhere. Some variations may be added by the individual speakers. They know what they are. But for NNS there must be some sort of point of reference. It's not like pronunciation is completely random although it allows for variation.
    Anyway, long live the CEPD!
    Best wishes,
    Silvia Cortese

    1. Thanks, Sylvia! I'll certainly pass your feedback on to my editors at CUP.

  4. Many people will get lot of benefits by reading this kind of informational stuff .Thank you so much for this .
    Online vocabulary

  5. Such valuable and relevant detail imparted by you, I am contented reading this information from this article. Keep blogging.
    English Practice App | English Learning App

  6. You asked about glottal stops and I don't see a reply yet. You acknowledge phonemic transcription, and the glottal stops of English aren't phonemes. They're components of preglottalized voiceless stops (including the voiceless affricate), so they're allophones governed by distrubutional rules.

    1. Thanks, Sidney. Yes, I'm aware that glottal stops are allophones. As they are so prevalent as inter- and post-vocalic variants of /t/ in some accents of English, however, it may be something we'd want to indicate more formally in CEPD, just like we indicate flapped /t/ in AmE. But that is something that would need to be discussed by the editors. As we are not currently being given an opportunity to do any update work to this dictionary - including adding new lexical items - I can't see that happening any time soon.