What the ‘ll’


I’ve written before about the name Purcell, and the evidence that indicates it must originally have been pronounced with first syllable stress, but it isn’t the only name ending in –ell to be stressed in this way: Marvell, Durrell, Cavell, Parnell, Angell, Mitchell, are all traditionally stressed by bearers of the name in the UK (I can’t answer for other places) on the first syllable. I know that Americans tend to pronounce the first two of these names with stress at the end: Mar’vell, Du’rrell, and I have heard an English professor of English Literature stress Andrew Marvell‘s name in this way – and I seem to remember that John Wells, in his blog some years ago, commented on this with surprise. Certainly the usage of both Gerald and Lawrence Durrell was ‘Durrell (initial syllable stress). Edith Cavell, the 1st World War nurse executed by the Germans as a spy is most frequently heard with final stress, but evidence collected by the BBC from the family is in favour of initial stress, and, at a service of commemoration in Norwich Cathedral, I have heard her supporters also use initial stress. The same is true for Parnell, surname of the 19th century Irish Protestant nationalist Charles Stewart Parnell: initial stress.

In all the cases bar one (Parnell), the etymology of these names, according to the Oxford Names Companion at least, shows that initial stress is the original pattern: Purcell (“swineherd”), Durrell (“hardy” – from French ‘dur’), Cavell (“bald” – ultimately from Latin ‘calvus’). The cases of Marvell, Angell and Mitchell (from “mickle” – meaning “great” or “large”) are self-explanatory. The exception, Parnell, is apparently a diminutive of the given name Petronilla, so second syllable stress would be understandable. Howell and Powell (= ‘ap Hywel’) are both derived from the Welsh name, and never, to my knowledge, stressed other than on the first syllable.

One of the most famous –ell names is Liddell – as in Liddell and Scott (Greek Lexicon), Alice Liddell (daughter of the above, and eponymous heroine of Alice’s Adventures in Wonderland), and Eric Liddell, hero of the film Chariots of Fire. I have never heard this name stressed other than on the first syllable.

It is a frequent feature of English language family names that a final consonant should be doubled, not only –l, perhaps as a way to distinguish it from the vocabulary word from which it is derived: Crabb, Mudd, Abbott, or simply as a doublet for the form with a single final consonant: Hewitt, Hewlett (both meaning “little Hugh”), Lockett, Waylett. The members of this last group are often diminutives. Where these names are polysyllabic, there is no reason to change the stress to the final syllable.

Another device used by some families is to add a final -e to the vocabulary word from which their name is derived: Blacke, Browne, Greene, Cliffe, Fowle(s), Groome, Rowe, Wilde.

Of course, before the days when spelling became standardised, those with a single and double final consonant letter were both in use, whether for the vocabulary word, where it exists, or the name, but the doubling seems to have persisted more readily for the names. Likewise for those names now distinguished from the vocabulary word by the ‘addition’ of a final –e: before the fixing of spellings, both forms were used for the vocabulary word and the name.


  1. Readers who’ve enjoyed reading these int·resting comments of Graham’s might like to see some similar remarks essentially on the name Marvell that I made in my Blog #110 on ‘One Way Pronunciations can Change’ at the 4th of July 2008 at http://www.yek.me.uk

  2. Another name I’ve never heard stressed on the final syllable is Russell. As with Liddell, I think the syllable structure and the double spelling of the middle consonants are a bit of a safeguard.

    This is surely the same mechanism that makes some people pronounce harass(ment) on the second syllable, namely that double spellings seem to “draw” the accent (or rather, that the accented “short” vowel often was given a double-consonant spelling when the orthography consolidated), together with the impression that these words are somehow French.

    I’m not sure the spellings of names as different from vocabulary words was typically a conscious choice. Proper names just aren’t as much subject to standard orthography. In addition, there are many names that actually have a different etymology and just merged phonetically with a vocabulary word or another name, so that there’s even less reason to use the same spelling. (Some bearers of the family name of Daniell go back to Norman D’Anneville, for example, others to the first name of Daniel.)

  3. I always assumed that stress on the second syllable of names like Marvell and Purcell derived from doubling the consonant, in a mistaken belief that the doubling served as a guide to emphasizng the second syllable.

  4. Re Lid(d)ell, I’m surprised that an ex-BBC man like yourself seems to have forgotten the announcer of yesteryear, Alvar Liddell (not sure of spelling), whose name was certainly stressed on the second syllable. Might he have been of Swedish origins? That would certainly account for the final-syllable stress.

    My own bête noir is the Scottish surname Lamont, which everybody except the former Chancellor (Lord) Norman Lamont pronounces as LAM-unt. In Gaelic it is MacLaomuinn – pronounced roughly Mac-limmin – which in turn gives rise to McClymont and McClement(s). The original word was the Old Norse “lagmaðr” – meaning “lawman”.

    The story I have heard about Lord Lamont, “peerie Norman” (little Norman), as he was known back home in Shetland, is that on going to Oxford University and cultivating Tory circles there he decided to give his proletarian Scottish surname an Anglo-Norman twist, with final-syllable stress imparting a pseudo-French flavour.


  5. Harry – You’re right: Alvar Lidell (one ‘d’) was of Swedish origin, which accounts for the stressing of his name. As for Norman Lamont, the Pronunciation Unit did ask him for his preference, and it was for stress at the end. I think most Scots feel about it the way you do! Among other names with a double final consonant that I didn’t mention are the composer Michael Tippett, and the playwright R C Sherriff. I’ve never heard either of those stressed on the second syllable.

  6. When working at Christ Church, where Liddell was Dean, I did hear one person pronounce the name with stress on the final syllable, but without good reason. We know how it was pronounced from the contemporary rhyme:

    I am the Dean, and this is Mrs Liddell,
    She plays the first, and I the second, fiddle.

    Our next door neighbours have the same surname, but I do not know how they pronounce it. My wife’s parents used to refer to a room at their church by the name of the Liddell Hall with stress on the second syllable until they encountered the rhyme. This may have been influenced by a church member with surname “Tickell”, which was definitely pronounced with the stress on the second syllable, though how far back in his ancestry that would have applied, who knows?

  7. The doubling of final letters (not just consonants) and addition of an otherwise unwarranted ‘e’ are used not only to distinguish names from common nouns as Graham points out, but also ‘content’ words from 2-letter ‘grammatical’ words since (excluding foreign imports) English spelling reserves 2-letter words for ‘grammatical’ items and interjections. Hence ‘inn’, ‘bee’, ‘too’, ‘bye’, etc where there is a ‘grammatical’ homophone. It also explains ‘odd’ not ‘od’, ‘egg’ not ‘eg’, ‘dye’ not ‘dy’, etc.
    I wrote a paper on this as a graduate student over 40 years ago and christened the phenomenon the Bamford Two-Letter Rule. I was laughed at.

  8. Alec – How do you cope with the exceptions: e.g. ‘ax’ (but not the British spelling ‘axe’), ‘ox’, ‘do’ and ‘go’?

  9. Oh, he’s a nitpicker, isn’t he? I don’t know how you regular contributors put up with him.
    Spelling ‘rules’ are very much toward the ‘tendencies’ end of the spectrum and ‘ox’ and, for transatlantic spellers, ‘ax’ are exceptions that prove the proverbial.
    But ‘do’ and ‘go’ have dual ‘content’ and ‘grammatical’ roles, the first as the auxiliary that appears in negatives, interrogatives and emphatics and the second in the periphrastic future ‘going to’.
    I did say the Bamford two-letter rule was laughed at, but not that much.

  10. Just for the record, Graham, as I have just stumbled across this blog, in my (extensive) family in Norfolk our name Durrell has always been pronounced with stress on the first syllable. But I have met people with the name spelled with single -r- (Durell) who put the stress on the second syllable.

  11. Thanks, Martin! I’m glad you found us, even if we had to trip you up to get you here.

Leave a Reply

Required fields are marked *.