Saturday, 4 February 2012

Concordia, Giglio, Stefano & Co.

[This week’s post was kindly written by Jack Windsor Lewis: many thanks, Jack!] 

Alex, you commented
 January 14, 2012 at 10:39 am

I’ve been watching BBC World News for a couple of hours now and they’re talking about the Italian cruise ship, Costa Concordia, which ran aground off the coast of Tuscany yesterday evening.

In this report by Andy Moore we hear that the island near to which the ship sank is called [ˈʒiːliəʊ], but no! Italians call it /ˈdʒiʎʎo/, with an AFFRICATE in the first syllable. …

And another newsreader, talking about the little town of Santo Stefano (...), called it [ˈsæntəʊ steˈfɑːnəʊ]. Italians always stress the first syllable in ‘Stefano’! 

I wasnt in the least surprised to hear such pronunciations. Despite the entries in LPD and CEPD, you're at least as likely to hear from even careful music presenters like BBC Radio 3's Rob Cowan the pronunciation /ʒiːli/ for the name of the great tenor Beniamino Gigli. I noticed that the BBC on-the-spot reporter Bethany Bell was saying /ʒiːliəʊ/ for the island even a day or so later. My article at §3.7.I.5 on my website ( pointed out some years ago how GB speakers are getting more and more vague about whether to use /ʤ/ or /ʒ/ except that for forren words they tend to prefer the latter.

As to Stefano, I pointed out in my Blog 107 on "Amphibrachising", if English speakers have the impression that words come from Romance languages, since in most such languages there is the much more frequent alternative in polysyllables of the amphibrach with the stress pattern [– `— –], they go for that. This is extremely familiar especially in large numbers of diminutives from albino and casino to mosquito and zucchini. Compare also items like bravado, desperado, libido, querido, tornado and torpedo. The tendency can be seen in eg the placename Cordova where, although all the various places so called throughout the Spanish-speaking world are accented Córdoba, all the half-dozen places so called in North America are given amphibrachic accentuation.

Many Italian, Spanish and other words and names are either often or regularly amphibrachicised by English-speakers including angora (now replaced as a city name by Ankara which is not so treated), Brindisi, Cagliari, Cyrano, Desdemona, Genoa, incognito, Lepanto, Lipari, Maritimo, mascara, Medici (tho the singular medico is usually okay), Modena, Monaco, Otranto, paprika, rococo, stigmata, Stromboli, Taranto, tombola, Trafalgar etc. The word oregano is amphibrachicised in Britain but not in America where Spanish is better known. More on this topic may be found on my website at 'Italian Words in Spoken English' Item 9 §4.


  1. Thanks for this, Jack/Alex... Jaleck?

    Obviously French is by far the "best known" foreign language to Brits, which I think explains their squeamishness about pronouncing "j" as dʒ. The effects of this go beyond Italian to e.g. Hindi, hence "Raj" and "Taj Mahal" with ʒ. Of course this British "knowledge" of French doesn't extend to stress, hence amphibroken Renáissance, Debússy, etc.

    Americans have Spanish as their "best known" foreign language, but they nonetheless do better than the Brits on French stress. So they not only say garáge, clichÉ, etc., but also refrain from amphibreakage, so we get Rénaissance and either Débussy or, even better, Debussý. They also have non-amphibroken Cýrano on the basis of its French origin.

    I must say I've never heard amphibroken Monáco on either side of the Atlantic.

  2. You may be right about stress, but in vowel quality BrE is often closer. For example, neither AmE /ˈdɛbjusi/ nor /ˌdɛbjuˈsi/ is a better approximation of the French original than BrE /dəˈbuːsi/. At least BrE gets the quality of the first vowel right.

    There is a similar argument with "renaissance", where again BrE gets the first vowel correct as a schwa.

  3. Obviously French is by far the "best known" foreign language to Brits, which I think explains their squeamishness about pronouncing "j" as dʒ. T

    Perhaps, but Americans seem equally squeamish, despite the fact that Spanish (which does not have ʒ) is better known there.

  4. I love how the transatlantic treatment of loans never fails to generate competitiveness!

    The reduced initial vowels in Brits' Debússy and Renáissance are mere accidental side-effects of the inaccurate stress pattern. Similarly, the better (but still not accurate) American stress pattern on Rénaissance generates an accidentally incorrect ə in the second syllable. (By the way, Brits are just as likely to use ɪ in the first syllable of Debussy, which isn't accurate.)

    The fact remains that Americans do French stress far better than the Brits, who have every reason to know better, while the Brits are a total disaster on Spanish, for which they have no excuse. So I'm afraid vp is on very thin ice when it comes to Brit/Yank differences regarding "j"! Americans generally turn Spanish "j" into h, which is pretty accurate. Brits may have just about been trained to use h in "fajitas", but most of them overcome their frenchy squeamishness about dʒ and inaccurately use it in "jalapeño" -- most of them also applying Great Vowel Shift inaccurately to the stressed "e" while they're at it, and ignoring the tilde which Americans pretty accurately turn into nj. Average Americans routinely handle names like "Jorge" [hórhe] and "La Jolla" [lə hójə], but I dread to think of the various ways in which average Brits would mangle these.

    Re vowels, I once wrote a paper on the rules which Yanks and Brits use in foreign vocab, but I no longer have a copy of it. Lindsey, G. (1990) Quantity and quality in British and American vowel systems, in S. Ramsaran (Ed.), Studies in the pronunciation of English (pp. 106-118). I may do a blog post on this at some point.

  5. So I'm afraid vp is on very thin ice when it comes to Brit/Yank differences regarding "j"

    I fail to see why. I was not talking about the grapheme J generally, but its correspondence with the sound /dʒ/, specifically the "hyperforeignism" of representing a non-English-origin /dʒ/ sound as /ʒ/ (as in, for example, "Taj Mahal", the example you gave). Based on my experience, that is at least as common in the US as the UK. Spanish orthographic J is irrelevant here, although I do agree that AmE generally nativizes Spanish loanwords in a manner closer to the original.

    I love how the transatlantic treatment of loans never fails to generate competitiveness!

    That's quite evident, since you have thrice claimed that AmE handles French stress "better" than BrE (which is itself an interesting claim, since French lacks lexical stress).

  6. On both sides of the Atlantic, turning dʒ into ʒ in foreign vocab seems to be an extension from French (where else?). Vp is quite right to imply that my best-known-foreign-language explanation is undermined if Americans turn Italian dʒ into ʒ as much as Brits. They may well do so; I'll listen out more.

    French isn't #1 foreign language for Americans, but their stressing of French vocab independently suggests the influence of French in their pronunciation of foreign words. (They sometimes inaccurately extend final stress to non-French words.)

    Whatever the domain of French stress is, it's final. Brits get that wrong, Yanks do better (sic). The clearest exception I know is m(o)ustache, which works the other way round!

    (And of course, when I first invoked "j", I should crucially have mentioned "g" before "i/e".)