I recently figured out how to view my unread Gmail, and was horrified to see how many links people have sent me have languished, apparently ignored and forgotten, because of my bad habit of letting them hang around until I need them. Here’s one the much-missed Paul Ogden sent me back in 2014 (!): Ampersand, An International Journal of General and Applied Linguistics.
Serving the breadth of the general and applied linguistics communities, Ampersand offers a highly-visible, open-access home for authors. An international, peer-reviewed journal, Ampersand welcomes submissions in applied and historical linguistics, phonetics, phonology, pragmatics, semantics, sociolinguistics and syntax. […] In response to the global thrust toward open source, open data and open access in science, Ampersand offers the opportunity for authors to make their research freely available to everyone, opening their work to a wider audience and increased readership.
Delving at random into the archives, I find “English language teacher development in a Russian university: Context, problems and implications” by Tatiana Rasskazova, Maria Guzikova, and Anthony Green, “English collocations: A novel approach to teaching the language’s last bastion” by Rafe S. Zaabalawi and Anthony M. Gould, and “Tweaa! – A Ghanaian interjection of ‘contempt’ in online political comments,” by Rachel Thompson; it looks like there’s lots of interesting stuff there, though I have no idea how well regarded the journal is by linguists. It’s too late to apologize to Paul, but I hereby issue a heartfelt “Sorry!” to all those who have sent me links and never heard back or seen them posted; hopefully they’ll be showing up belatedly in posts to come.
OK, I’ll bite. Pity about the paywall on the Ghanaian interjection …
FWIW [tɕʏɪaa] for “tweaa” is presumably wrong: the tw is [tɕᶣ], as in the name of the language, Twi, which sounds like “chee” to an uninstructed L1 English speaker. J G Christaller’s extraordinarily good Twi grammar of 1875 (of which I possess FHW Migeod’s copy) actually calls the language “Tshi”, though he rightly describes the initial consonant sound in loco as “a mixture of ‘ch’ in ‘church’ and w.”)
I would guess the word has something to do with ɔtweá, glossed in Christaller’s Twi dictionary as “dog, bitch; a mean worthless fellow, good-for-nothing fellow, wretch.” Christaller gives the interjection (twéaa) itself, too: “interj expressing the utmost disregard or contempt.”
Well, mistaking an /y/-glide for /Y/ isn’t that big a mistake.
It’s not a glide, but a coarticulation. Huge mistake. They should apologise at once, and promise never to do it again.
(It may be my IPA: basically tw is a voiceless palatal affricate with lip rounding. The English ch j are often somewhat labialised too, which is why to English speakers Twi tw dw sound very like ch j.)
Hey, it’s downloadable after all. Stupid me. I’ll take a look.
OK. I looked. Und bin so klug als wie zuvor …
It’s not a glide, but a coarticulation.
So it is. And yet /kʷ/ seems to become /kw/ with comparative ease, as in Latin > Italian.
It may yet happen with Twi (but not yet …)
Hausa kw represents /kʷ/ rather than /kw/ too; Kusaal adopts it as kp [k͡p], as in bakpae “week” from Hausa bakwai “seven.”
If you want to hear a completely unrounded sibilant, check out Navajo. The first few times I tried to imitate “left”, I couldn’t stop my lips from rounding!
Do you analyze Italian as having a /w/ that only occurs in /kːw/ and /gːw/? Or do you analyze uo as /wo/ to add one more environment?
Do you analyze Italian as having a /w/ that only occurs in /kːw/ and /gːw/? Or do you analyze uo as /wo/ to add one more environment?
Ma, “equilibrio”?
OK, the short ones, too.
Well… /wɔ/ phonotactically patterns like a vocalic nucleus rather than a CV sequence — unsurprisingly, given its origin from (and synchronical alternation with) earlier /ɔ/. For instance, it selects the prevocalic allomorphs of the article: l’uomo [lˈwɔːmo], gli uomini [ʎˈwɔːmini] (nowadays mostly [ʎiˈwɔːmini], by a kind of spelling pronunciation).
However, I think it is better to posit a phoneme /w/ (albeit of somehow restricted distribution), since it otherwise patterns like a consonant, and does not alternate with [u] — not even in /wɔ/ (unless you have a Neapolitan accent, in which case pronunciations such as [uˈoːmo] are possible). Conversely, underlying /u/ can become [w] before another vowel: continuare often has [-nˈwaː-] in casual speech, but it can be [-nu.ˈaː-] in more careful pronunciation. But uomo or quando cannot have [u.V].
And /w/ is needed anyway for loanwords, where it does definitely pattern like a consonant: il weekend, i weekend. You do find spellings like l’whisky in older writing, presumably after the analogy of native l’uomo, l’uovo, etc., but it always struck me as unnatural — I definitely say il whisky.
That last point is particularly interesting because French famously treats initial [j] as a vowel, as in les yeux -[zj]-.
Given that French is also perfectly capable of treating zero as a consonant in liaison, it doesn’t necessarily follow that this is a synchronic fact about French phonemics.
Yes, I forgot to add that all the above more or less applies to /i/ and /j/ as well, mutatis mutandis.
Come to think of it, Italian does not have a lot of words with initial /jɛ/ < Lat. (h)ĕ- — in fact the only one that comes to mind is ieri ‘yesterday’, which as an adverb does not often carry the article.
Actually, for initial /jV/, of whatever origin, there seems to be more oscillation: /j/ is increasingly treated as fully consonantal, i.e. lo ieri (from diphtongized (h)ĕ-), lo iato (from Lat. hĭātus) are now preferred to l’ieri [lˈjɛːri], l’iato [liˈaːto]. Note the use of lo, not il, which means that this /j/ is doubled between vowels.
There are still cases with underlying /j/ not alternating with [i]. These are either obvious loanwords (lo yacht) or less obvious ones (lo iettatore, from VLat. *iectare, via a Southern dialect that did not turn #j- to #ʤ-; la iella, Roman dialect form of obscure origin, in fact not unfrequently spelled with ).
… “with i“, I suppose.
“…with j“. Sorry!
@jc
No he meant with j, “la jella”. He doesn’t say, but “lo jettatore” is also possible (may be an obsolete spelling, though).
@Davids E & M:
French also treats /w/ as a vowel for the purposes of liaison, e.g., [lezwazo], [lwazo] for “oiseau.” Same for /ɥ/, with [lɥitʁ], [lezɥitʁ] for “huître.” It might be most parsimonious to say that the trigger for liaison consonants is a highly sonorous first segment rather than a vowel tout court, or something to that effect.
French oi is pretty clearly a diphthong – the o doesn’t count toward the length of consonant clusters. (And it’s not quite [w] phonetically either, though it strongly labializes the preceding consonant.)
That, and likewise for ui, is in the paper on the Romanian diphthongs that we discussed recently.
The best case for a /w/ in French is équation. It does not quite sound like é-quoi-tion.
For comparison purposes (all the links have audio files):
https://en.wiktionary.org/wiki/uomo#Pronunciation
https://en.wiktionary.org/wiki/tuo#Pronunciation
https://en.wiktionary.org/wiki/huomenta#Pronunciation
https://en.wiktionary.org/wiki/huonosti#Pronunciation
(Finnish -uo-, -yö-, and -ie- have developed from proto-Finnic -oo-, -öö-, and -ee-.)