Mark Liberman had a post at the Log quoting a correspondent as follows:

I read your article on the alphabet olympics yesterday and followed one of the links, and then one of its links, and so on. I was merrily traipsing thru the internet when I came upon a page that threw me: “The Rules and Misrules of English Spelling“.
The note on “th” (note (f)) gives a list of words with the “this” sound (what I’d call “voiced th” — ð rather than θ) that includes the word “with”. I was surprised — I have always used unvoiced as the pronunciation of that word, and had never noticed anyone doing otherwise. Sure, voicing gets *added* sometimes due to context, but surely unvoiced is the target — right? Apparently wrong. My Pocket Oxford gives only the voiced pronunciation, and my Houghton Mifflin Canadian gives the voiced version first, as does my New Lexicon Websters. The two pronunciation sites I found online also gave voiced pronunciations.
I asked my wife to pronounce the word slowly and carefully, and she likewise gave an unvoiced pronunciation, and was surprised that anyone aimed for the other (tho’ she did point out that Bono has a buzzy version when he sings “with or without you”). (I grew up in Nova Scotia, and my wife grew up in southern Ontario.) OK, so I’ve got a non-standard (or less standard) pronunciation — it’s not the only one I have. I’m interested in what the distribution of this variant is, but I’m having a hard time finding it online.

I thought “Yes, I’ve heard people use a voiceless final in that word”; I checked with my own wife, and what do you know, she had a voiceless final herself. Well, today Mark posted a followup citing John Wells’s phonetic blog to the effect that 84% of Americans use a voiceless final (/wɪθ/), only 16% sharing my voiced /wɪð/. (In the UK, the proportions are reversed: /wɪð/ 85%, /wɪθ/ 15%—though /wɪθ/ is heavily favored in Scotland.) Eighty-four percent! Rarely have I been so astonished to find myself in a small minority (though I’m used to that situation in general).
Incidentally, John Wells is annoyed that people aren’t using his Longman Pronunciation Dictionary, but the damn thing costs $42.57. As Mark says, wouldn’t it be nice if Pearson made it available online? But they may feel that not enough people would pay a fee to use it to make it worth the trouble.


  1. Glad I’m not the only one who was surprised how few people use voiced th in with. Until I read the post, I believed that everyone used voiced th in that word unless there is some phonotactic reason to unvoice it.

  2. I suppose we can attribute this to the two sounds sharing orthographic th; combined with the low functional load of the distinction, most speakers aren’t even aware there’re two separate phonemes.
    It’s hard to imagine a similar variation between say /s/ and /z/ in a high frequency function word going unnoticed.
    Though many people would be surprised that of has /v/, so maybe I’m wrong…

  3. I take that back: if some people said /ʌv/ and others /ʌf/, I’m pretty sure it would be commented on, and eye-dialect spellings ov and off (or uv and uff or something…) would occur.

  4. It doesn’t surprise me at all that most people from the USA deploy a voiceless th in with and most in England don’t. Cockney “wiv” wouldn’t really work if the norm were voiceless, and it’s certainly voiced in my own, RP English. I think of voiceless in the British Isles as coming from Ireland (an Irish accent being more US sounding).
    The people who don’t voice the th in with: do they never voice it or does it depend on the following word?

  5. For me, it’s usually but not always unvoiced. I can’t find a consistent pattern:
    With – usually unvoiced?
    Within – usually voiced?
    Without – can go either way?
    Thin – usually unvoiced, though other word-initial ths are usually voiced.
    There aren’t any minimal pairs of th vs. th are there?

  6. Without – voiced
    Outwith – unvoiced.

  7. The first audio at the log seems to be saying “weth”.

  8. Also unvoiced (for me):
    Firth, Forth, furth, fourth.

  9. There is, actually, a sort of minimal pair: “with” versus “withe” (a tough supple twig, esp. of willow).
    Since the th of “with” is at the end of a word, where it is subject to being devoiced and lenited, probably the distinction for most people who do distinguish a voiced from a voiceless th in this word, is the length and quality of the preceding vowel.
    My own “with” has “voiced th”, which is really voiceless, with lengthened and inglided lax “i”. But I often hear the “voiceless” variety, also, especially in the prepositional verb “come with” used intransitively.

  10. It doesn’t surprise me at all that most people from the USA deploy a voiceless th in with and most in England don’t.
    Well, it shouldn’t surprise anyone that most people in England have a voiced th, since that’s the historical form. But that shouldn’t stop you from being surprised about the situation in the US. I mean, I’m a certified American who’s heard American speech all his life, and I’m surprised. When did this happen, I wonder? Since dictionaries largely still prefer the voiced form, it must be pretty recent.

  11. As another certified American, I am also shocked. It never crossed my mind that you can pronounce “with” with an unvoiced final consonant. I can’t recall ever noticing an American speak that way. I’ll have to listen more carefully in the future.

  12. J.W. Brewer says:

    I would be interested in seeing the underlying data for the 84/16 claim (which is said to be something called a “preference poll” — I don’t think it’s phrased as an unequivocal claim about actual use). I too find that level of supposed overwhelming support for unvoiced surprising — maybe it’s bad data, or maybe there’s a regional distribution such that I’m personally less exposed to this majority practice, or maybe as suggested above because the voiced/unvoiced distinction here isn’t used very often to distinguish minimal pairs of words our ears aren’t attuned to notice it as much as we notice voiced/unvoiced contrasts for other consonants.

  13. The alternation of voiced and unvoiced th is predictable by rules, but the rules are complicated and appeal heavily to non-phonological notions, so minimal pairs can and do exist. Here’s my best current shot at formulating the rules:
    1) Loan words never contain voiced th, even when borrowed from languages that have either [ð] (like Spanish padre, comrade < camarada, etc.) or /ð/ (like Modern Greek skordalia ‘Greek-style aioli’). This rule overrides all following rules. There may be very old loan words to which this does not apply, but I don’t know any.
    2) Morpheme-initial th falls into two categories:
    2a) The function words the, they, their, them, this, these, that, those, there, thither, thence, then, thus, than, thou, thy, thine, thee (and perhaps a few more) are voiced. Note that although < all though, so its th is voiced too.
    2b) All other initial th, as in thin, think, thick, Thor, thorough, thief is unvoiced. Through is a function word, but it is unvoiced because it is just a shortened form of thorough. Known minimal pair: thigh/thy.
    3) Th which is both morpheme-medial and intervocalic is voiced: father, weather, lather, etc. Exception: brothel is traditionally unvoiced, though many people now pronounce it voiced by analogy (I do). Intervocalic rth is also voiced, even in rhotic dialects: northern, worthy, etc. All other morpheme-medial th is unvoiced (if any; I can’t think of any examples). No known minimal pairs, but near-minimal earthy/worthy, where the latter is no longer felt as worth+y.
    4) Morpheme-final th falls into three categories:
    4a) With is unvoiced in Scotland, Ireland, and America, but voiced in most other places. Australia is apparently variable in this respect. Outwith is unvoiced in Scotland and unknown elsewhere.
    4b) Morpheme-final th that was once intervocalic before the fall of final e and the reduction of plural -es mostly retains its voicing. The nouns baths, booths, oaths, paths, sheaths, truths, wreaths, youths and sometimes moths end in unvoiced th, but are voiced in the plural (taking /z/ as the plural ending). This voicing is completely lost in the U.S., maybe in other locations too, and some people in England have it in only some nouns. For the same reason, denominal verbs like breathe < breath, bathe < bath are consistently voiced. Lathe < lath also belongs to this family. There are probably others. Bathing may be < bathe or bath (v), and is voiced or unvoiced accordingly. Known minimal pairs: lo(a)th/loathe, mouth (n)/mouth (v), wreath/wreathe, sheath/sheathe, sooth/soothe, teeth/teethe.
    (All these were once intervocalic and are exactly analogous to f/v alternation in leaves, wives etc. and s/z alternation in houses and for some people by analogy spouses. None of this kind of fricative voicing applies in Scotland, at least not in varieties close to Scots.)
    4c) Other morpheme-final th is unvoiced, as in pith, death.

  14. Charles Perry says:

    I grew up voiced and am still find the voiceless pronunciation odd. Was there a time when gangsters started being depicted as saying “wit youse” instead of “wid youse”? (The gangsta pronunciation is evidently voiceless, as in “I wanna get witchoo.”

  15. Charles Perry says:

    Hey, I used voiced th in baths, booths, oaths, paths, truths, wreaths and youths (but not, for some reason, oaths and sheaths), so that voicing is not completely lost in the U.S.

  16. J.W. Brewer says:

    Are you callin’ me un-American, John Cowan? Is this the new phonetic McCarthyism? Because subject to the usual dangers of introspection, I think I’ve been generally voicing the th in with my entire native-AmEng-speaker life and it’s only in the last few days that I’ve first become aware of claims (which I am not yet accepting the empirical veracity of) that this puts me in a minority among AmEng speakers.

  17. John,
    I voice singular “booth” (e.g. telephone-booth) and I believe that’s a common (if not standard) pronunciation in the UK. The surname “Booth”, too.

  18. Charles Perry: I stand corrected as to that detail, but not as to the general principle: the noun plurals are irregular, and therefore subject to loss. In my speech, at least, and in that recorded in U.S. dictionaries, all are lost.
    Apparently the traditional accent of New York (and, I’d guess, at least some other Eastern cities, probably not Philadelphia) have voiced with too.
    I should also say that -ern in northern, southern, eastern, western is a cranberry morph, so these words are now monomorphemic, despite their transparent relationship to north, south, east, west respectively.
    Michael Peverett: I had vaguely heard that booth was voiced even in the singular in the U.K., but couldn’t confirm it. Well, I’ll be, yet another individual exception like brothel. Booth is also a fine example of those “very old borrowings” I mentioned in point 1, as it was borrowed from Old Danish before interdental fricatives were lost in the mainland North Germanic languages.
    Keep those exceptions coming in, folks.

  19. Australian: I used the voiceless variety for many years before discovering (in my teens or twenties, I think) that the voiced variety is supposed to be correct.

  20. This is a but of a baader-meinhoff moment for me. I’m taking a Intro to Linguistics class and transcribed “with” as voiced on a quiz a few weeks ago, which was initially marked incorrect. I was shocked to realize that most of my east-coast classmates weren’t voicing it. Now I’m just waiting to get home to the Bay Area to see how most of my old friends say it–I hope the voiced version isn’t just an idiosyncrasy.

  21. Cowan: “3) Th which is both morpheme-medial and intervocalic is voiced: father, weather, lather, etc. Exception: brothel …”
    Ethel, ethyl, methyl, ether(eal), ethos, ethics, Ethan, Ethiopia, catholic, apothecary, catheter

  22. I should also say that -ern in northern, southern, eastern, western is a cranberry morph, so these words are now monomorphemic, despite their transparent relationship to north, south, east, west respectively.
    I guess that depends on your analysis and criteria. I personally don’t find them monomorphemic. Each can be analysed into a morpheme (adjective, direction of the compass), followed by the non-productive morpheme ‘-ern’. I’m wondering how you would handle ‘northerly’, etc.

  23. I’m just saying I wasn’t surprised, not that I wasn’t surprised because I know of a whole lot of evidence. And I’m a certified USian too, I just have a British accent.
    Booth as in “General” Booth or Connie Booth (American) are both voiced (by me & I’d guess by most Englishpersons).
    Is there anyone who doesn’t voice “smooth”?
    I used to work with a guy from Arkansas who said “roof” to rhyme with “woof” and “roofs” to rhyme with “woofs”, whereas I say “roof” to rhyme with “youth” and “rooves” for the plural. So there.

  24. J.W. Brewer says:

    Crown: rhyming “roof” with “woof” is reasonably common in a number of non-prestige varieties of AmEng. I grew up over a thousand miles ENE of Arkansas, and “roof” was one of four words I can recall my 8th grade English teacher using as shibboleths to distinguish “incorrect” dialect pronunciation from standard/prestige pronunciation (with rhymes-with-woof on the wrong side of the divide). But I can’t recall ever hearing “yoof” for “youth” in an AmEng context although i take it from eye-dialect written representations I’ve seen that it must be common in some (stigmatized?) varieties of BrEng.

  25. I’m not sure of the origin to “yoof”, but I think it’s probably a 1960s comment on cockney Rolling Stones-type emerging culture and may come from “youth club” which was a publicly-funded place where yooves could meet in the evenings (I think).
    My workmate from Arkansas was also a Yale grad. I’m guessing his accent wasn’t totally unprestigious. He was quite emphatic about his pronunciation, I remember, sounding almost like a dog.

  26. I’m guessing that a youth club had fluorescent lighting, a linoleum-tiled concrete floor, two ping-pong tables and many orange plastic stackable chairs.

  27. No one in our certifiable American family voices “with” — and I think that explains why our teenage son always thought Bono was singing “Wither without you.”

  28. With is unvoiced in Scotland, Ireland, and America
    Well, in “prestige” dialects in America, i.e. the Northeast, it is still voiced.
    I just tested my wife, from Philadelphia, and she pronounces “with” unvoiced as well. This is terrifying – how long has this been going on? I feel like the protagonist in Invasion of the Body Snatchers – “They’re here already! You’re next! You’re next!”

  29. It had never occurred to me that “roof” needn’t rhyme with “woof” though of course I’ve met many people who pronounce that leading “r” as “w” anyway.
    I have met people who pronounce “roof” as “roove” and “grease” as “greaze” and, come to think of it, they had the trait, rare in Scotland, of using the voiced pronunciation of “with”. Hoodlums, I suppose.

  30. I say both woof[lax] (like “look”) and woof[tense] (like “moon”). It’s tense in “warp and woof”, but lax in “my dog goes woof, woof”. And I say both roof[lax] and roof[tense]. It’s lax in “the roof of my mouth”, but tense in “so mad he went through the roof”, and could be either in “fix my roof”. I’m from northwest Ohio.

  31. Born WV 1948, now live in KY. Say “with” with voicing. Have only noticed the unvoiced variety on NPR. Also say “roof” lax and “root” tense (is there some sort of complementary distribution pattern for those words)?
    Ah well, as Howard Cruse said, make woof not warp.

  32. I just tested my wife, from Philadelphia, and she pronounces “with” unvoiced as well. This is terrifying – how long has this been going on? I feel like the protagonist in Invasion of the Body Snatchers – “They’re here already! You’re next! You’re next!”
    My reaction exactly!
    I used to work with a guy from Arkansas who said “roof” to rhyme with “woof”
    I have Arkie roots myself (father’s side) and I think I rhyme it with “woof” about half the time. I have a number of linguistic uncertainties like that.

  33. mollymooly says:

    @Greg Lee: your 3)s are covered by 1): “Loan words never contain voiced th … This rule overrides all following rules.”
    However: algorithm, logarithm, rhythm, zither seem to be exceptions (although Wells’ dictionary gives a voiceless US variant for zither).

  34. Other exceptions:
    (Webster’s Third New International Dictionary of the English Language gives two alternative pronunciations \ˈnädə, ˈnäðə\; the original transcription uses an underlined th)
    (/aɪˈsteðvɒd/, along with /aı`stedfɒd/)

  35. Rog: (is there some sort of complementary distribution pattern for those words)?
    That’s what I’m wondering. I think John Cowan will know.
    I have a number of linguistic uncertainties like that.
    Me too. I think some people are more susceptible to acquiring bits of new accents than others. While living in the US I leaned to say “bin” for been about half the time (before that I only said “bean”, like most Englishpersons).

  36. @ mollymooly “Loan words never contain voiced th …”
    Oh, I see. Then how about “Netherlands”?

  37. I’m English-born — ‘with’ is certainly voiced but ‘booth’ (whether the name or the almost defunct telephone installation) is unvoiced. Rhyming ‘booth’ with ‘smooth’ sounds very odd to me. I grew up in southern England but my parents were from the north. To my ear, unvoiced ‘booth’ fit wells with a Yorkshire accent.

  38. And as a possible exception to Cowan’s 3) (since my first set was mistaken), “gotham”.

  39. J.W. Brewer says:

    My old AHD lists both the tense and lax (in that order) pronunciations of roof, but my 8th grade teacher (whose views I’m not endorsing as scientific or anything . . .) had no doubt that the first was proper and the second not. She would have linked them to “route” (in the “get your kicks on Route 66″ sense) just b/c it was another of the shibboleths – the correct/prestige pronunciation being homophonous with “root” and the “wrong” low-class variant being homophonous with “rout.”

  40. J.W. Brewer says:

    Greg Lee: is “Netherlands” really a loanword or just a calque of “Nederlanden” (and/or a somewhat archaic-on-its-own synonym for Low Countries, which calques Pays-Bas)? “Nether” as a freestanding word now sounds a bit old-fashioned, but one does see it in e.g. the euphemism “nether regions.”

  41. Greg Lee: Gotham is indeed an exception. The village in Nottinghamshire is now spelled Gatham.
    Dearieme: There’s no GOOSE/FOOT split in Scotland anyway.
    Mollymooly: I knew about those exceptions to rule 1, but forgot about them. I suspect the fact that the following sound is a syllabic consonant not present in the original language has something to do with it.
    I was actually mocked for using the GOOSE instead of the FOOT vowel in roof when I moved to the New Jersey town where I spent most of my childhood. I ignored the abuse / And stuck with GOOSE. When I told my wife about this several years ago, she immediately said that /rʊf/ is what a dog says.

  42. J. W. Brewer asked “is “Netherlands” really a loanword or just a calque of “Nederlanden” …
    Yes, both. Looking at the Wikipedia definition for “calque” below, I conclude that a calque is a kind of loan:
    In linguistics, a calque (/kælk/) or loan translation is a word or phrase borrowed from another language by literal, word-for-word (Latin: “verbum pro verbo”) or root-for-root translation.
    I suspect that something else is going on here. Noting that the Cowan rules seem to be, synchronically, impossible to interpret for a naive speaker, who is not going to know whether a form is borrowed or not, I think there might be a synchronic rule that voices “th” after a short stressed vowel and before unstressed “er”, and that guides speakers who see written “th” and have to figure out whether to pronounce it voiced.

  43. After reading the LL article, I am now too self-conscious of how I say “with”. I sampled my wife and 5-year-old daughter. Wife pronounces it unvoiced. Daughter pronounces it with a voiced consonant somewhere between ‘dh’ and ‘v’. Both are native-born Americans, living in the San Francisco Bay Area; my wife grew up in Iowa.

  44. Proud Dog Owner: My dog can talk. Tell him, boy. What’s on top of a house?
    Dog: Roof!
    Stranger: Get out of here!
    PDO: No, no, listen: What’s sandpaper like:
    D: Rough!
    S: Scram!
    PDO: No, no, hang on. Tell him, who’s the greatest baseball player of all time?
    D: Ruth!
    S: Beat it. And take your damn dog with you!
    D (on the way out): Maybe I shoulda said Hank Aaron …

  45. I seem to voice the end of “with” most of the time, but the more attention I give to the matter the more unsure I am about that “most of the time”. Certainly I have long been aware of the unvoiced version, and possibly I have a lingering sense that it is “right” or “better”.

  46. Here are more forms that occur to me in support of my proposal above that “th” becomes voiced between stressed short vowel and unstressed “er”:
    (Cotton) Mather
    (Orrie) Cather
    (Sally) Struthers
    Of course there are many more ordinary words that have voiced “th” in the relevant position, but proper or less usual words are more to the point, if we want to figure out present day speakers understand their language.

  47. I suspect you’re going to be shocked to find that Mrs Ø doesn’t voice it. The conclusion I’m coming to is that women in the US don’t voice “with”, just as Norwegian & German women quite often take an inward hiss of breath instead of saying “yes”.
    Having done extensive research I can now reveal that John Cleese says unvoiced “Booth”, but voiced “Cleese”.

  48. AJP – yes but John Cleese’s family name was originally “Cheese”.
    J.W. Brewer: I can’t recall ever hearing “yoof” for “youth” in an AmEng context although i take it from eye-dialect written representations I’ve seen that it must be common in some (stigmatized?) varieties of BrEng.
    Yes, in Cockney/lower-end Estuary: the modern “Jafaican” pronunciation among under-35s in London would be “yoot”, as in “da yoot”.

  49. What sort of person changes their name from Cheese to Cleese? You didn’t see Harvey Milk’s parents trying to cover up the dairy-product connection. And so half arsed, why not change it to Baker or Constantinides?

  50. Not youts?
    Youf is common in AAVE too. Think the Jay-Z line “Brand new convertibles/I’m so roofless (ruthless)”

  51. It might have been mere corruption rather than a conscious change.

  52. J.W. Brewer says:

    I’m not well-positioned to be getting aural confirmation on my work computer, but I associate “roofless” for “ruthless” with NWA’s “Gangsta Gangsta” (where it rhymes with “toofless”) from approx. a quarter century back. I suppose if that’s really a standard AAVE-ism it might generalize to yoof, but I can’t personally recall having heard that and (correctly or not) took “toofless” and thus the following “roofless” to be not a standard AAVE-ism but a jocular imitation of how someone’s speech might be altered if they’d just (as would follow from the context of the narrative) been punched repeatedly in the mouth by the young-and-not-yet-mellowed-out Dr. Dre.

  53. A note on my own pronunciation of initial /ð/, which has puzzled me for years, as I’ve always known it’s different from the standard. I have spent some time now thinking about the mechanism.
    I pronounce both /θ/ and /ð/ with the common interdental variant and my /θ/ is an exemplary interdental. But somewhere along the way I have picked up the other variation for /ð/, the plosive /d/.
    But when I try to pronounce the target /d/, I have a problem, because the interdental tongue position means there is aperture via the very back teeth.
    It seems that to approximate the silent plosive onset of /d/ I restrict airflow at the glottis (creaky voice). So I have a (very) creaky-voiced fricative onset (~20ms) followed by plosive release, or affricate if the voice happens to kick in just before the plosion.
    The creak restricts airflow so much that I had a lot of trouble figuring out over the past hour(!) exactly where the sound was coming from.
    The IPA for this is hard to write in this size but here ɡoes [ð̰̟͡d̟], that is a creaky /ð/ tie-barred with /d/, all advanced (viz. interdental).
    I suspect the target phoneme /d/ is due to a father with Cockney features and this is why it’s only associated with word-initial, voiced /ð/.
    Don’t know if this is interesting, never heard tell of this variation before though I have been doing it all my life.

  54. “F” for final theta occurs in AAVE, but I don’t know if it’s “standard.” Earlier in that verse, Ice Cube pronounces “youth” with what I think is a “th,” but I’m pretty sure Youth has an “f” in Dead Prez’s Behind Enemy Lines and definitely in Zion I’s The Bay. Freddie Gibbs says “boof” in 187 Proof. Jay-z says “troof” in Otis. Boots says “wealf” in The Coup’s The Name Game. Eazy-E says “rufless” in Ruthless Villain. I hope all these are accurate.
    Sorry for the random songs, I’m bad at finding relevant lyrics. And there are a lot of counter-examples, so it’s not a hard and fast rule. It’s also hard to be sure when it’s used in a song because -ooth words are often rhymed with -oof words, like in C.R.E.A.M. “Leave it up to me while I be living proof/To kick the truth/To the young black youth(youf?)” I have the nonscientific impression that it may be more common in certain words (“boof”) than others (never heard “wif,” instead “wit”).

  55. “I mean, I’m a certified American who’s heard American speech all his life, and I’m surprised. When did this happen, I wonder? Since dictionaries largely still prefer the voiced form, it must be pretty recent.”
    It has to be old enough for dialect forms like “what’s up witchoo? to derive from it.

  56. J.W. Brewer says:

    Interestingly, Gangsta Gangsta itself has “wit” for with (not “wif” as to the rarity of which I am happy to defer to Joe R’s superior knowledge) in a later verse, but that’s arguably forced for the sake of the rhyme (with “shit”).

  57. I once studied with an African American whose English was entirely standard except for final /θ/ > /f/. Thus he said “maff” for math. I never asked him about it, but I expect he preserved this one signal of ethnic identification.

  58. for “this one” read “this as one”

  59. Just found a great example of how variable this is. Was listening to 2 Chainz’s Birthday Song (yeah, I know), and in the chorus and throughout it’s “birtday,” but in the outro he says “And it’s your birfday, baby.”
    And a final example, from Biggie’s Juicy “Birfdays was the worst days/ Now we sip champagne when we thirsty.”
    Okay, sorry for spamming. No more.

  60. Greg Lee: Your -ther /-ðər/ rule definitely does not work across a morpheme boundary, as truther ‘one who does not believe in the usual theory of 9/11′ and birther ‘one who does not believe that Obama was born in the U.S.’ exemplify. Trying out such nonce words as clother (cloth+er, not clothe+er), deather, eighther, faither, fifther, growther, healther, pither (one who piths frogs?), seventher, sixther cause my internal machinery to report /θ/ as well.
    Creating that list caused me to look into the verb badmouth, which is listed in m-w.com, NID3, AHD4, and RHD2 with both voiced and unvoiced pronunciations, presumably because some derive it from the verb mouth, some (like me) from the noun. Alas, ODO gives no BrE pronunciation. What say the Hattics of the five sub-countries of Rightpondia (viz. Londonia, Eboracia, Bagpipia, N’Iron, and Quaint)?

Speak Your Mind