Talk:Italian phonology/Archive 2

Archive 1

Archive 2

Archive 3

Vowel length in Italian

The author proposes: "However, vowels in stressed open syllables are long (except when word-final)."

As a non-linguist, I have the impression that Italian simply does not know the quantity of vowels as it is known from, say, the Germanic languages, or maybe French, or Czech, Latvian or Finnish. Stressed vowels, not just in open syllables, tend to be "drawled" but this seems to be a phenomenon concomitant upon word-stress, and restricted to situations when the word is logically or emotively stressed, or perhaps pronounced as a stand-alone example, or some such. In the beginning of 'Pinocchio' (http://ia700406.us.archive.org/2/items/avventure_pinocchio_librivox/avventurepinocchio_01_collodi_64kb.mp3) I must admit I can't hear any long vowels no matter how hard I try in non-prominent words such as 'registrazione', 'sono', 'capitolo primo', 'Collodi' and most of the others. This is very unlike what we call vowel length in languages which really have it, like the afore-mentioned ones.. . 89.74.217.56 (talk) 23:14, 7 January 2011 (UTC) Wojciech Żełaniec

You may be confusing phonemic and solely phonetic vowel length, and expecting that long vowels will be of equivalent length across languages. To take Finnish as an example, /Vː/ and /V/ can and do determine minimal pairs. In Italian length is allophonic, with length occurring with regularity only in stressed open syllable. The /a/ of pane and the /a/ of panino do differ in length, with the first noticeably longer than the second. The length itself may sound -- and probably is, depending on the speakers -- "half long" as compared to Finnish, yet clearly longer than the open syllable stressed vowel of Spanish. Compare a Central Italian's rendition of Roma with a Finn's Rooma and a Spaniard's Roma. 47.32.20.133 (talk) 19:49, 23 August 2018 (UTC)

No I am not confusing those two things, I am well-aware of the difference. Still less do I expect that long vowels should be of equal length across languages; I've had way too much exposure to different languages for that. If, as you said, vowel-length in Italian is merely allophonic, what role does it play in that language (why are Italians so consistent in observing it (answer: they aren't!)) and why should we be reminded of it by means of the ugly colon? I'd question your statement, besides, that it occurs regularly in open syllables. It does occur, but neither regularly (I am not referring to that 'endearing' kind of mocking Italian by saying 'ro:ma', 'pa:ne' 'vi:no' etc. all the time), nor just in open syllables: I hear my Lombardian friends drawl their stressed vowels (but not regularly!, only when logically or emotionally emphasised, or before a pausa, but even this is neither regular nor predictable) even in closed syllables, e.g. "me:zzo" or "no:cciolo" or what not... No, I think the existence of the vowel-length in Italian is a myth, read-into Italian by scholars who had a 'real' (whatever this be...) vowel-length in their own languages. Besides, it sounds so cute to mock an Italian by saying 'mie:lo e:stero zucchera:to" or "Gi:gi Amoro:so", it's a bit like mocking a German by interspersing German words with superfluous glottal stops or a Swede by 'singing' one's mock-Swedish phrases or such-like. This said, I don't want to deny that word-stress in Italian may, and probably does, consist in occasionally lengthening the vowel of the to-be-stressed syllable. But the same is true of my native Polish, amongst other tongues, which no-one in his right mind would accuse of having vowel-length (like Italian, though, it does have consonant-length or -gemination; English and German haven't it, only (British English) vowel length, so does Czech and Slovak, Swedish, Finnish, Norwegian have both.) Well, lascia:te mi canta:re, la chita:rra in ma:no, lascia:te mi canta:re, so:no Italia:no (ve:ro). Ain't that cute? 134.95.92.16 (talk) Wojciech Żełaniec

No, more silly than cute. And there are no long vowels in chitarra because there are no stressed open syllables. You're quite right, though, that vowel lengthening is not heard regularly in the Italian spoken in Lombardy, and lengthening can occur where it doesn't in so-called standard Italian. That's true of most of the North. Otherwise, Uguzzoni captures it very economically, speaking of stressed vowels: una vocale lunga è seguita da consonante breve (V:+C), mentre una vocale breve è seguita da consonante lunga (V+C:). Find an Italian to say "Che fanno a Fano?", and you'll hear the difference, probably even from most northerners in that case. Not surprisingly, it's more complex in the prosody of extended unguarded discourse (which you would know if you had checked out Bertinetto & Loporcaro (2005) in the article's references), but it's not untrue. If you develop a genuine interest in Italian phonology, you can get a very good start with Schmid, Stephan. 1999. Fonetica e fonologia dell' italiano. Torino: Paravia Scriptorium. 47.32.20.133 (talk) 02:44, 29 October 2018 (UTC)

Lasciate mi cantare con la chitarra in mano is a song, and singing is not like speaking. Singing teachers usually advise their student to prolong every vowel or other syllable nucleus as much as possible because they are the sounds that carry pitch and dynamics best. Love —LiliCharlie (talk) 03:05, 29 October 2018 (UTC)

Of course silly, I was using that to mock an Italian-mocker who prolongs every stressed vowel and thinks that's 'cute', or that's the way an 'italia:no" should sound. As I said above, I consider it as obvious that Italian stressed vowels (not at the end of a word, like in 'liberta`') and not only in open syllables are quite often somewhat longer than non-stressed ones. Phonetically, that is. But, I'd think, this is just part of what their being stressed, word-stressed, consists in. In Polish (my native tongue), where the penultima is stressed somewhat more consistently than in Italian (in some variants ONLY the penultima can be stressed), being stressed is a mixture of such factors as: pitch, loudness and length. Which of the three (there are possibly some others) is the dominating in a given word in a given utterance depends on various rather difficult-to-predict factors. One of these is the age and the origin of the speaker, and there are people in whose speech the length (i.e. being longer than non-stressed vowels) is the dominant factor. I sometimes speak this way too (e.g. after longer stays in Italy...) but generally older citizens from the formerly Polish eastern territories (now Lithuania, White Russia, Ukraine) have this tendency: to prolong stressed vowels, or rather: to make them stressed by prolonging them. Yet no-one has so far proposed that contemporary Polish (even regional) has vowel-quantity. Is the situation in Italian not like that, too? If I say "che fanno a Fano" while stressing "Fa-" by pronouncing it more energetically or louder, but not lengthening the [a] in it, can I then be misunderstood or non-understood? Or will it be immediately obvious that I am not an Italian native speaker, may the rest of it sound as Italian as ever? Part of the problem seems to be that if you ask an Italian to pronounce a word in isolation, slowly and clearly, in a 'didactic' way, he or she will reliably prolong the stressed vowel. But in 'real life' speech he will do it only if there is logical or emotional emphasis to be put on the relevant word. If 'Fano' is at stake, it'll be 'che fanno a Fa:::no?' (as distinct from anywhere else), but if 'Fano' is presupposed and the activity is at stake, it'll be 'A Fano, che fanno?' with no lengthening of 'Fa-', and probably with 'che' in a higher pitch. But these are matters of word-stress and phrase intonation, not vowel-length. In similar situations, even vowels in closed syllables will be prolonged. 134.95.92.16 (talk) Wojciech Żełaniec

As for being misunderstood, read the article: "In Italian there is no phonemic distinction between long and short vowels, but vowels in stressed open syllables, unless word-final, are long." That implicitly takes you back to your observation when you opened this thread: in Italian vowel length doesn't work the way it does in e.g. Finnish. Quite right. Word-internal vowel length is phonemic in Finnish; it's not phonemic in Italian. -- If 'Fano' is at stake, it'll be 'che fanno a Fa:::no?' (as distinct from anywhere else), but if 'Fano' is presupposed and the activity is at stake, it'll be 'A Fano, che fanno?' with no lengthening of 'Fa-'. Not normally, no. The /a/ of Fano in the first one could be superlong, but without knowing the individual speaker's quirks, there's no reason to expect that it will be. In the second case, preposing normally topicalizes, so that with Fano preposed there's actually a little more likelihood of a superlong /a/ in Fano, and none that the /a/ will be short. -- Basta. These talk pages are not the place for Italian Phonology 101 lessons. Check out Schmid and the more serious of the article's references, and the bibliographies therein. (You can even test yourself. Once you have a little bit of understanding, you'll be able to explain why Italian has buono but bontà. Once you get the basics down pat, you'll be able to explain not only why Italian has ferro, not fierro, but also, more interestingly, why festa, not fiesta.) [imbokkalluːpo] 47.32.20.133 (talk) 12:57, 29 October 2018 (UTC)

I am not -- at least not here -- interested in Italian Phonology lessons; them there ferro-fierro issues, Historical Phonology issues I suppose, aren't quite unknown to me, either -- but you see, as a guy who occasionally (much too often) buys books by those who have they PhD's in Italian and other languages, I'd like to get things convincingly explained to me. Coming from a language which has no vowel-length at all and some (much less than does Italian, though) consonant gemination, I had to learn 'the hard way' what it is to understand and to understandably speak a language which has vowel length and no consonant gemination. But these hard-won skills are not straighforwardly transferable onto languages like Italian, of which scholars whose native languages have vowel length say it has it, too. (And scholars of other mother-tongues usually don't.) In other words, a 'long vowel' seems to be quite a different thing in Italian than it is in German, Czech or Finnish. But you guys somehow get away with glossing over this issue. Much to the 'benefit' of us punters.... Well. it's clearly enough if we buy your books... 153.19.31.79 (talk) Wojciech Żełaniec

In other words, a 'long vowel' seems to be quite a different thing in Italian than it is in German, Czech or Finnish. One more time: Yes. There's no chance for word-internal vowel length to be distinctive in Italian. But you guys somehow get away with glossing over this issue. No one who knows what s/he's doing glosses over it. Any comparative study will make the differences quite explicit. 47.32.20.133 (talk)

In very technical papers -- might well be, you're probably right. But in dictionaries, textbooks, the WP etc. the phonemic transcription of Italian words often includes the ominous ':' sign. E.g. this: "Rome (Italian: Roma [ˈroːma] (About this sound listen); Latin: Roma [ˈroːma]) is the capital city of Italy", in the WP. Forgive me if I appear stubborn, but I think it's wrong to suggest that the stressed vowel in a non-final open syllable is always /albeit non-phonemically/ a *long* vowel. A person who took this to heart and put it to practice would sound ridiculous -- or as mocking Italian. The truth is that it sometimes is, and sometimes isn't. In Latin it might have been [ˈroːma] or even /ˈroːma/. In Italian it's neither, I feel. My hypothesis: the lengthenings, wherever they occur, are part of what word-stress in Italian consists in or, in other words, is realised as. So marking the word stress in the phonetic transcription should be enough /presupposed that the reader knows how to stress words in Italian.../ 134.95.92.16 (talk) Wojciech Żełaniec

the phonemic transcription of Italian words often includes the ominous ':' sign. E.g. this: "Rome (Italian: Roma [ˈroːma] That's phonetic transcription. /phoneme/ [phone] is the convention. /ˈroma/ → [ˈroːma] is shorthand for "the phonemic structure /ˈroma/ is realized phonetically as [ˈroːma]." I think that's sort of what you're trying to say in your last two sentences immediately above, i.e. given a structure and its stress, a speaker in command of the phonology can pronounce it in a normal way. 47.32.20.133 (talk) 21:26, 29 October 2018 (UTC)

Yes, I am sorry, it should have been 'the phoneTic transcription', that was what I meant and mistyped. Of course, I am aware of the difference between '/.../' and '[....]'. But I still see '[ˈroːma]' as dangerous, to wit, because it can misguide the reader to pronounce the city's name with a long [o] always, consistently, as she would with the German word 'Buch' (book) phonetically transcribed as '[bu:x]'. In German, it's a relatively long(er) vowel always, no matter what the context, the pragmatics, or any other relevant circumstances are like. That's what I call a 'robust vowel quantity'. In Italian by contrast -- well, if my untutored ears have heard right for the past 50-something years -- it's not always ['ro:ma], sometimes just ['roma], and this depends on the quirks of the individual to a considerable extent. And some other times ['ro:::ma] or worse still. If the relative length of the stressed vowel is part of what its being stressed consists in, in a given utterance, then in another utterance being stressed may consist rather in being pronounced louder, more energetically, or with a higher pitch instead, rather than lengthening -- following some rules that escape us (or just me) yet. Or -- who knows -- maybe the lengthening of the stressed vowel is NOT part of what its being stressed is realised as, but is situated on a different, say emotional, rhetorical or whatnot, level of language as actually spoken (parole?). To repeat: if I am under the impression that 'Roma' should always and consistently be realised as ['ro:ma], 'mano' as ['ma:no], etc., and this is what this transcription suggests, then I might end up speaking a rather strange-sounding version of Italian. Or am I wrong on this 'un? 134.95.92.16 (talk) Wojciech Żełaniec

Pretty much, yeah. Very determinedly so, it seems. 47.32.20.133 (talk) 01:21, 30 October 2018 (UTC)

Here: http://etcweb.princeton.edu/cgi-bin/dante/almagest/cgi/audio2?poem=inf&canto=1

Nel mezzo del cammin di nostra vita

mi ritrovai per una selva oscura

ché la diritta via era smarrita.

Ahi quanto a dir qual era è cosa dura

esta selva selvaggia e aspra e forte

che nel pensier rinova la paura

Tant' è amara che poco è più morte;

ma per trattar del ben ch'i' vi trovai,

dirò de l'altre cose ch'i' v'ho scorte.

I honestly hear only 'oscura' and (the second) 'era', and perhaps 'amara' as (slightly) drawled, as far as the stressed syllable vowel is concerned. In all three cases, the lengthening is quite like what you can expect from a Polish person (if it had been Polish, of course) from a certain region, even though no-one claims that contemporary Polish has vowel quantity. I don't hear it in 'vita' or 'diritta' or 'smarrita' or 'cosa' and so on. I should add that I can speak and understand several languages which have real vowel quantity, such as German or Swedish (which has consonant quantity or gemination, in addition) and I usually both hear and realise the long vowels correctly in these languages. Italian -- no, they do lengthen their stressed vowels sometimes, and sometimes they don't but this is not at all (I'd insist) anything like the vowel quantity of Czech, Finnish, Latvian or even French. This trait (i.e. occasional lengthening) is exploited by mockers and parodists, and certainly you're well advised to drawl your accented vowels sometimes if you want to sound authentic (plus a certain whiney tone of voice), but a systemic part of language ... I dunno. But clearly, too many academic careers would be blemished if this imho erroneous assumption were withdrawn. 153.19.31.79 (talk) Wojciech Żełaniec

I hear it very clearly in primo, vita, trovai, definitely not d(i)ritta, but via and smarrita, yes, ditto cosa, dura, and so on. The /i/ of smarrita is a tiny bit less long that might be expected, and it sounds a little odd because of that. When he moves fast and doesn't stress, as in era at one point, no length. I don't know why the difference in perception, but you seem to be fixated on comparing to other languages rather than tuning your ear to Italian on its own, and that may be part of the problem. You also keep mentioning drawling; you may be expecting a good bit more length than is normal -- when the length is natural, it doesn't really sound anything like the foreigner's sing-song exaggeration that you've caricatured before. (If you would bother to read you'd see that some of those in the most successful academic careers are actually trying to sort this out, and agree with you more than you think.) 47.32.20.133 (talk) 13:30, 29 October 2018 (UTC)

ok, so your ear is obviously finer than mine, all the better! I can't hear the length in 'primo', 'vita' or 'trovai' /well, a diphthong as such is long, innit?/ at all, nor in 'cosa', 'smarrita' and so on. To be more precise: while in Italian word stress involves, like in so many other languages, a slight lengthening of the stressed syllable's vowel, and this component is in Italian more prominent than it is in Polish (where loudness and pitch seem more important) or Spanish, I can't hear any such lengthening in these words as could possibly generate the illusion of a truly (whatever that means) long vowel. In 'oscura' by contrast -- yes. In 'vita' -- no. I will have a look at Bertinetto & Loporcaro, thank you. Maybe it'll restore my faith in this kind of study... As for tuning my ear to Italian: well, I know what vowel length is, e.g. from German /e.g. 'all' vs. 'Aal', eel/, a perfect minimal pair, this is phonemic, robust, hard-working, while in Italian it's allophonic at best, flimsy and invites silly caricatures /pizza, mafia, espresso, and 'Gi:gi Amoro:so/ 153.19.31.79 (talk) Wojciech Żełaniec

German /e.g. 'all' vs. 'Aal', eel/, a perfect minimal pair, this is phonemic, robust, hard-working, while in Italian it's allophonic at best For the nth time, yes. Phonetic only, a classic case of totally automatic allophony. There are no minimal pairs possible in Italian of the sort (C)V:C(V) vs (C)VC(V), and there is no reason whatsoever to assume that, regardless of its phonological status, duration of vowel length in one language will match duration in another. flimsy and invites silly caricatures I have no idea what flimsy means in this context. Length is variable? If that's what you mean, of course it is. Ma che fai? can be said in many ways. If you're especially frustrated with whatever Tizio is doing, the /a/ may be super long. Or it could be a very short clipped remonstrative version in which length is all but absent. Totally expected variation, normal. (Ask speakers of North American English if they ever "drop their aitches". They'll almost always say no. Then get them to repeat He has a hat on his head several times. You'll soon arrive at a clear hierarchy of likelihood that the /h/ will drop.) As for caricature, the choice to do that is up to the speaker, as it is with the peculiarities of any language.

I entirely agree that Italians prolong and over-prolong their vowels for various expressive reasons. Or more broadly: all kinds of pragmatic reasons. This is obvious. And they do it very well, most of the time... Yet, in a language like German the vowel of "Aal" (eel) is always relatively longer than the adjacent 'short' vowels, no matter what the relevant pragmatics is. That's what I call 'robust' vowel-length, is it just called 'phonemic' technically? As for caricatures, I think this excessive drawling fits in well with various such properties as being overly emotional, theatrical, etc. deplorably still quite commonly ascribed to Italians; which is why I dislike this kind of caricature. As an aside: English and German (vowel length present) scholarly publications usually mention vowel length in Italian, while Polish (no vowel length) don't. People quite often like to 'discover' in a foreign language something from theirs, or overlook something theirs hasn't. I once knew an educated Chinese who believed German had tones, or a Swede who believed Swiss German had tonal accents (like Swedish). Again, a German language course in Swedish tells us 'gosse' (Swedish for 'boy', consonant gemination) is pronounced exactly like German 'Gosse' (gutter, no consonant gemination, which doesn't exist in German), which it isn't. Or that German guy in Italy who kept saying 'latte', he needed some milk, with a single 't', like 'Latte' in German, meaning 'lath' and no-one understood him. 153.19.31.79 (talk) Wojciech Żełaniec

is it just called 'phonemic' technically? No. If it's called phonemic, it's because it's phonemic. 47.32.20.133 (talk) 17:38, 29 October 2018 (UTC)

Sorry, that question of mine was ungrammatical; I meant: 'is it precisely that which is technically called 'phonemic'?' To which your answer was 'yes', wasn't it? 153.19.31.79 (talk) Wojciech Żełaniec

A phonologist has just arrived from another galaxy, lands in Italy. S/he discovers that [ˈaːno] and [ˈanːo] are judged by native speakers to be different words, so there must be a phonemic difference between the two. Test to find out what. (...) S/he goes through the variations and discovers that [ˈaːno] and [ˈano] are not different words ("but only foreigners say [ˈano]"). Ditto [ˈanːo] and [ˈaːnːo] (not different words, but "[ˈaːnːo] sounds a little strange"). After some more testing (fato - fatto, pala - palla, etc.) and noticing alternations (pane, panificio; capace, capacità), she concludes that there is a phonemic contrast between "double consonants" and single ones, and that stressed vowels in a syllable not closed by consonant are phonetically long, but there is no phonemic contrast between long and short vowels. S/he hypothesizes that the phonemic structures of [ˈaːno] and [ˈanːo] are /ˈano/ and /ˈanno/ respectively. 47.32.20.133 (talk) 19:56, 30 October 2018 (UTC)

OK, but my point has been from the outset and still is that even the most native of Italian speakers _not always_ say ['a:no] or ['fa:to], but sometimes, too, ['ano] or ['fato], so it's not just foreigners. This is the Italian I am familiar with. If someone thinks it's always and invariably ['a:no] she will start speaking a funny Italian. To give an example of a somewhat analogous issue: I have an Italian friend who is in the habit of speaking slowly to very slowly whatever language he happens to be speaking (he is a rather pensive human being). Now he picked up the American voicing of [t] between vowels, (a disanalogy with our case: from what he heard, not from transcription), but he did not pick up the rule that your [t] 's godda be the more voiced the faster you are speaking. As a result, he would say, tempo lentissimo: 'A is bedda than B', 'that's a different madder', and the like, which sounds rather funny -- as does too 'Trahno, the capidal of Ondario' or 'senns' (for 'sentence') -- if you speak very slowly, as is his _naturel_. (In fact 'senns' for 'sentence', in a lentissimo utterance, can lead to misunderstandings.) Now imagine that an American English dictionary for foreigners indicated this type of allegro pronunciations as the standard or the only pronunciations quite generally. Would slow speakers who took this to heart not start sounding funny? I don't at all deny that it is often ['a:no] or ['fa:to], or that it's often (most of the time, in fact) 'bedda' or 'Trahno' in N. America, but I am uneasy about raising such pronunciations to the level of a norm. BTW isn't "to hypothesize" something of an Italianism in English ('ipotizzare'), or else a non-equivalent of 'to put forward a hypothesis', meaning, not the former, but rather to 'speculate', 'fantasize' or such like? 2A02:A312:C43D:DC00:D92C:8EB4:9368:3B5D (talk) Wojciech Żełaniec

... and his 'twenty' is [twɛ̃ɾi], sort of, spoken tempo lento; sounds delightfully entertaining. 2A02:A312:C43D:DC00:D92C:8EB4:9368:3B5D (talk) Wojciech Żełaniec

No. 47.32.20.133 (talk) 16:12, 31 October 2018 (UTC)

Here's Gassman doing Canto 1: https://www.youtube.com/watch?v=aBGq11ODudA Listen to his intro as well; you'll hear lots of more natural length there (Italian length, not German, Polish, Finnish...) 47.32.20.133 (talk) 14:54, 29 October 2018 (UTC)

Amazing, beautiful, Italian is a very nice-sounding language. I start suspecting this whole 'vowel length' phenomenon in Italian is entirely _sui generis_, not comparable with the more pedestrian vowel lengths in German, Czech, Swedish etc. It deserves separate treatment on all levels of description134.95.92.16 (talk) Wojciech Żełaniec

I start suspecting this whole 'vowel length' phenomenon in Italian is entirely _sui generis_, not comparable with the more pedestrian vowel lengths in German, Czech, Swedish etc. LOL! ¡Por fin! 47.32.20.133 (talk) 01:21, 30 October 2018 (UTC)

What I start suspecting is that [ˈvɔi̯t͡ɕɛx ʐɛˈwaɲɛt͡s]'s sui generis means that it is futile to deal with Italian vowel length in the way this article currently does, for only insiders are able to understand this presentation. Love —LiliCharlie (talk) 02:18, 30 October 2018 (UTC)

This is what this article claims:

"In Italian there is no phonemic distinction between long and short vowels, but vowels in stressed open syllables, unless word-final, are long."

And this is what the referenced source claims:

"Stress is phonemic. Stressed syllables show a tendency to lengthen the consonant in coda, if there is one; if there is no consonant, they lengthen the vowel. (This phenomenon, known as ‘phonetic lengthening’, is controversial.) But word-final stressed vowels remain short."

What is the exact reason a phenomenon which the source calls "a tendency" and "controversial" is presented as an established fact here? Love —LiliCharlie (talk) 10:01, 30 October 2018 (UTC)

The source gives no indication of what is meant by tendency, nor of what is controversial about vowel length in open stressed syllables. Perhaps your question can be answered by whoever it was who selected that article as a source. If detail is needed in this article, something like Bertinetto and Loporcaro's observation could be added, so as to introduce the concept of variability of length in unguarded connected speech, then an example: given that "all (non-emphasised) primary stresses but the utterance-final one are considerably weakened in the speech chain" (143), ... 47.32.20.133 (talk) 14:15, 30 October 2018 (UTC)

A very low profile answer: tendency -- sometimes perceptibly lengthened, sometimes not; controversial -- likely to give rise to discussions like this one. I shall check out the B & L book, I promise, 153.19.31.79 (talk) Wojciech Żełaniec

It's an article. Just click on the pdf link. 47.32.20.133 (talk) 19:56, 30 October 2018 (UTC)

That there is a 'tendency' to lengthening the stressed vowels (and yes, even in closed syllables, if my ears are any guide) I have no quarrel with. What I do have a problem with is erecting this 'tendency' into a hard phonetic fact, by means of such transcriptions as ['ro:ma], alongside and on a par with such ones as e.g. this one: 'Maximilian Karl Emil Weber (/ˈveɪbər/, German: [ˈveːbɐ]'. [ˈveːbɐ] will be [ˈveːbɐ], in German, regardless. The Italian ['ro:ma] can be ['roma] or ['ro:::ma] depending on the pragmatic context, the quirks of the speaker and such like. This is, among other things, what makes this tendency controversial, as this thread has made sufficiently evident, methinks. But it also makes it so interesting... 153.19.31.79 (talk) Wojciech Żełaniec

What I do have a problem with is erecting this 'tendency' into a hard phonetic fact, by means of such transcriptions as ['ro:ma], alongside and on a par with such ones as e.g. this one: 'Maximilian Karl Emil Weber (/ˈveɪbər/, German: [ˈveːbɐ]'. [ˈveːbɐ] will be [ˈveːbɐ], in German, regardless. 1) No claim or equivalence whatsoever is made about German in the description of Italian. A description of language x is a description of language x, not of a,b,c...y, z. . 2) Variation is normal. No brief description is going to include every possible detail. Even an extended description of American English phonology is unlikely to mention that "I am going to tell you..." is realized by some speakers, in some registers, as [ˈãmõˈtʰɛˑjˑə]. 47.32.20.133 (talk) 19:56, 30 October 2018 (UTC)

Please have some mercy with us or'n'ry punters (=buyers/users of books written by professional linguists, readers of WP entries on phonology et., in this case). We have learnt on our own the complex API symbolism, the difference between -emic and -etic, langue and parole and so on.... and now you're telling us that '[...]' transcriptions mean quite different things depending on which languages they are being applied to? If it reads '[ˈveːbɐ]' it means a German will always say '[ˈveːbɐ]', give or take the absolute value of vowel duration in a given utterance -- but that's because it's German, while, when it's Italian, ['ro:ma] means that some Italians will say sometimes ['ro:ma] and at other times, or other Italians -- something else. Because 'no equivalence claim made'? This is very exacting, I think.... 2A02:A312:C43D:DC00:8043:6748:3CB0:70E8 (talk) Wojciech Żełaniec

If it reads '[ˈveːbɐ]' it means a German will always say '[ˈveːbɐ]' Record a trio of Germans half full of beer bickering over some point of Weber's work. If they are observed to maintain every Weber as [ˈveːbɐ] in a long heated discussion (be sure not to conflate tense/lax with actual length), you've got material for an excellent article. The journal Language Variation and Change would be happy to publish it. (Friendly tip to save time, frustration and eventual embarrassment: before you start, read everything you can find on phonological variation in general and variation in vowels in German specifically.) 47.32.20.133 (talk) 16:12, 31 October 2018 (UTC)

That's a poor example. 3 Germans with 3 different temperaments and 3 different speech paces, bickering (and this can be very brutal, amongst Germans), beer. I have made it sufficiently clear, or so I hope, in my previous chimings-in, that I am lucidly clear about the merely RELATIVE character of this whole short vowel long vowel distinction, haven't I? A short vowel in one utterance can be (even in the same speaker) in absolute numbers decidedly longer than a long vowel in another. And in a situation like that of your three Weber-critics that's even extremely likely. Well... Why not consider once again the two examples I provided earlier, the preface to Collodi and Dante? In the latter, sorry, while I can here ['oscu:ra] (sort of) I can't hear ['vi:ta]. Before a pause Italians are likelier to prolong their non-final stressed vowels than they are in mid-sentence, this I know from years of exposure, am I wrong? 2A02:A312:C43D:DC00:D92C:8EB4:9368:3B5D (talk) Wojciech Żełaniec

It's an excellent example. If it reads '[ˈveːbɐ]' it means a German will always say '[ˈveːbɐ] is either true or not true. To find out which, put it to the test. A phonologist with no knowledge of German would guess that there would be variation in the vowel length of Weber in the scenario described. -- No offense, but the formula incomprehension + cantankerousness renders discussion tiresome and pointless. 47.32.20.133 (talk) 18:04, 31 October 2018 (UTC)

'If it reads '[ˈveːbɐ]' it means a German will always say '[ˈveːbɐ] is either true or not true. To find out which, put it to the test. A phonologist with no knowledge of German would guess that there would be variation in the vowel length of Weber in the scenario described.' Yes, no doubt, you're quite right -- I'd never think of denying that. But I thought I had made clear that I concede this. I've heard so many discussions like the one you describe in the real world!... But I also thought that '[ˈveːbɐ] stays [ˈveːbɐ] no matter what' is true in the sense that the [e:] in every of the name's occurrences is slightly longer (German long vowels, unlike e.g. Latvian ones, are not all that terribly long) than the corresponding short vowel (here: let's abstract from the tense-lax issue) in more or less immediately adjacent words in the same speaker's mouth. Of course, if Hans is speaking much faster than Fritz, his [e:] in [ˈveːbɐ] will be possibly shorter than Fritz', and even than Fritz' [e] or [ɛ] in the same heated debate. Or, if he is speaking accelerando or rallentando, his [e:] may be shorter than his OWN [e] a moment earlier or later on. No doubt about it. But, always, or perhaps not absolutely always (one should abstain from such generalisations in the sublunar world) but in most not-extremely-unusual situations it will still be longer than the [e]s immediately preceding or following. (Yes, I am aware too of all these imprecisions involved here: 'slightly' longer, 'a moment earlier' etc.) This again, if I may so boringly remark, I have gleaned from decade-long observation and study of languages with generally recognised and non-controversial vowel-quantity. Sorry for this stubbornness -- old men are often like that, you see -- but this is what I (and certainly not just me) hear. Cantankerousness -- sorry for unintentionally having given this impression; I'd dream of getting a few things explained convincingly by some of those whose studies and academic careers I have for so long been co-contributing to financing, by my taxes and my purchases. So my questions were not meant to cause anybody any nuisance -- just 'Fragen eines lesenden Arbeiters' to speak with Bert Brecht, questions from a reading working-classes bloke. Thank you for bearing with me for so long 2A02:A312:C43D:DC00:D92C:8EB4:9368:3B5D (talk) Wojciech Żełaniec

'in a long heated discussion (be sure not to conflate tense/lax with actual length),' thank you Prof, but I can't quite honestly admit not to have been aware of this issue... Let's imagine then 3 Czechs with Czech beer in them and some heated debate on Janáček between them... In Czech long vowels and short are identical (except for [i]), and occur both in stressed and in non-stressed syllables alike... That's a very solid vowel-quantity status, one Italians can only dream of... 2A02:A312:C43D:DC00:D92C:8EB4:9368:3B5D (talk) Wojciech Żełaniec

Why an 'established fact'. I used to believe these two 'theories': 1) people whose native idioms to have a real(ish) vowel length (English is probably on the -ish side of the line) naturally and excusably tend to 'hear' vowel length wherever there is an ever so small tendency to lengthen some vowels. 2) All three highest-prestige European languages, English, French and German, do have a real(ish) vowel length (though in none of them is it as neat, hard-working and free from non-quantity-relevant admixtures as it is in Czech or Finnish or Latvian). So it's in a way flattering for Italian to count it as a vowel-quantity language as well. And easier to accomplish (lengthenings are far more frequent and perceptible) than e.g. for Polish, where there is too a certain tendency to lengthen stressed vowels. These are two very 'folksy' theories, no doubt... 153.19.31.79 (talk) Wojciech Żełaniec

Yes, Professor Żełaniec, as a Sinologist I suspect that if Modern Standard Mandarin Chinese (MSMC) or Standard Cantonese had a high prestige in the Italian-speaking world people might start to always transcribe pitch-related phenomena in Italian. Of course (at least) all continuant sounds have length and all voiced sounds have pitch. However if no rationale is given why some phenomena that are mere tendencies are always transcribed like hard facts while others are completely neglected, the method seems to lack some of the characteristics that science after Descartes is supposed to have. Love —LiliCharlie (talk) 20:35, 30 October 2018 (UTC)

'However if no rationale is given why some phenomena that are mere tendencies are always transcribed like hard facts while others are completely neglected' -- one such rationale (if 'rationale' be indeed the right word here) is that the tendency at issue is particularly prominent in the ears of those whose national cultures have highest-prestige languages with a similar 'tendency' or more, and whose scholarly cultures have made and go on making major contributions to Phonology and similar fields of study (btw the phonemic-phonetic distinction is reported to have been first made by, not an Anglo, a Franco or a Teutono, but by a Pole, Jan Niecisław Ignacy Baudouin de Courtenay; however, he was technically a Russian then, as Poland did not exist in the XIX century, so all is well) have. I don't think Italians themselves like being flattered by the ascription of vowel length to their 'lingua di sì"; it's rather the others, methinketh, who believe to do them a favour in this way. So not directly transferably onto Chinese. All of this -- horribly unscientific, I know, and cynical to boot, and yet ... bare of EVERY particle of truth? 2A02:A312:C43D:DC00:D92C:8EB4:9368:3B5D (talk) Wojciech Żełaniec —Preceding undated comment added 16:59, 31 October 2018 (UTC)

Yes. That's why I said those two were two folksy 'theories'. Without even a tiny grain of truth in them, full stop. ... Or perhaps a question mark... ? Amicus Cartesius, sed ... BTW: Here is a German dictionary's entry 'Roma':

https://de.pons.com/%C3%BCbersetzung?q=roma&l=deit&in=it&lf=it -- you never see such a thing in a dictionary whose publisher's main address is in a country with a language with no vowel quantity... Of course this is al unscientific and "honny soyt qui mal y pense", I know ... 2A02:A312:C43D:DC00:8043:6748:3CB0:70E8 (talk) Wojciech Żełaniec

Interestingly, unlike the German PONS dictionary, the Italian pronunciation dictionary dipi online does not mark vowel length for Roma or any other entry.

PONS gives phonetic transcription. As indicated in the Breve guida on line, Canepari gives phonemic transcription in DiPi. 47.32.20.133 (talk) 16:12, 31 October 2018 (UTC)

What also fascinates me is the further inconsistency that although the referenced source first mentions that "[s]tressed syllables show a tendency to lengthen the consonant in coda" this part of the sentence is completely ignored in this article's transcriptions: There is a [ˈrɔs.po] but no [ˈrɔsː.po], as well as many more analogous cases where the coda of a stressed syllable happens to be a consonant. Looks like somebody failed to notice that Rogers & d'Arcangeli fromulated a "coda length tendency rule" and not merely a "coda vowel length tendency rule." Love —LiliCharlie (talk) 09:54, 31 October 2018 (UTC)

There are various issues here. First off, what they said was this: Stressed syllables show a tendency to lengthen the consonant in coda, if there is one; if there is no consonant, they lengthen the vowel. The problem with the statement is the vague waffling 'tendency'. In actual fact, lengthening of coda consonants is a rather weak tendency, applied -- if at all -- most saliently to /n/ and /r/ (as demonstrated in their transcriptions of one speaker, a rather peculiar one, as evidenced by a couple of her renditions of, e.g., the geminate and stressed vowel of mantello, markedly Roman). [ˈrɔsː.po] is not impossible, but highly unlikely. As for the second major issue, you make a good point (sort of, implicitly): Rogers and d'Arcangeli's little article is not intended as a description of Italian phonology, and should not be used as such. It's clearly labeled ILLUSTRATIONS OF THE IPA, according to JIPA's own description of the long series: "Illustrations of the IPA are concise accounts of the phonetic structure of different languages using the Association’s International Phonetic Alphabet." Concise, not exhaustive, complete, detailed... illustrations of IPA. 47.32.20.133 (talk) 16:12, 31 October 2018 (UTC)

Well, http://www.dipionline.it, one of my favourite ones, is of Italian making, isn't it? Neither Anglo-Saxon nor Teutonic. Re the consonant coda issue I'll have a further question, layder. This proves only that this whole issue of 'Italian stressed syllables' is still far from ready to be presented as self-contained, complacent wisdom in Italian Phonology 101 courses. Which makes it all the more interesting, doesn't it?2A02:A312:C43D:DC00:D92C:8EB4:9368:3B5D (talk) Wojciech Żełaniec

This, too, is instructive: http://operaclick.com/forum/viewtopic.php?t=20507&p=650315 read Messaggio da mascherpa » 29 giu 2015 10:57 re the name of L. Janáček, as far as the complex connections between vowel length and word stress in Italian are concerned. 153.19.31.79 (talk) Wojciech Żełaniec

I have, finally, read the paper by Loporcaro and Bertinetto and ---what does it say? It says: 'Stressed vowels are lengthened in word-internal open syllables when they occur at the end of the intonational phrase (thus including isolated words) or under emphasis: cf. casa ["ka...sa] ‘house’ vs. cassa ["kas...a] ‘chest’ and casetta [ka"set...a] ‘little house’. Contrary to wide-spread opinion, this lengthening process is thus far from being a categorical word-level phenomenon,'

I am not sure I understand what that means, but it sounds like ... well... like what a lengthening is in a language with real vowel quantity, am I wrong?

'as observed in Bertinetto (1981), Landi & Savy (1996) and McCrary (2003), and as confirmed in the corpus-based study carried out by Dell’Aglio et al. (2002). The exact phonetic implementation of the stress-conditioned lengthening process is, in any case, prosodically governed even at the word level (Marotta 1985). Thus, although we mark lengthening in our transcriptions, even within stretches of connected speech, this merely indicates the pronunciation appropriate to isolated words.'

Precisely! Gotcha! Most dictionaries, textbooks and other publications for non-linguists do not have the epistemological decency of saying: 'Ladies and GEntlemen, this pronunciation that we here indicate is just that of isolated words.' Small wonder, then, that those who take those good-for-isolated-words-only transcriptions at face value start talking funny Italian. It's like telling people 'Toronto' is ['tra:nou] in American/Canadian English (but not adding: if you are speaking reasonably fast, that is).

Actually, it's like telling them that Toronto is pronounced [tʰəˈrɑˑntʰoʊ]. Which, assuming the transcription is accurate, is true. But it's not the whole truth -- as anyone with the least bit of common sense would suspect upon just a bit of reflection on variation (diaphasic, diatopic, diastratic) in his/her own language, even without hearing American 'Did you eat yet?' as two syllables ([d͡ʒitt͡ʃɛt). 47.32.20.133 (talk) 03:30, 1 November 2018 (UTC)

I was referring to my Italian friend, mentioned above, and his allegro forms in English while speaking lentissimo. But I agree your example is better ('bedda'), more to the point, because like ['ro:ma] or ['fa:to] it takes as its point of departure a 'canonical' transcription. (With the perhaps not quite insignificant difference that '[tʰəˈrɑˑntʰoʊ]' is used most everywhere, while ['ro:ma] mostly (exclusively?) in anglophone and germanophone countries.)

Millions and millions of Italians use ['ro:ma] every day. 47.32.20.133 (talk) 17:52, 1 November 2018 (UTC)

Sorry, I meant the transcriptions using colons. In few non-Anglo and no-Teutono dictionaries, textbooks and the like have I ever seen such transcriptions. Not here: http://dizionari.corriere.it/dizionario_italiano/D/dizionario.shtml, nor here: https://www.dizionario-italiano.it/dizionario-italiano.php?parola=dizionario, nor even here: http://www.wordreference.com/iten/dizionario,

There's no more need to tell an Italian that the stressed vowel in Roma is long or the /n/ in con Carlo is pronounced [ŋ] than there is to tell an anglophone American to aspirate the first /t/ of potato, but flap the second one. It's automatic for those speakers. Sabatini-Coletti is written for a very general public of Italians, is not a pronouncing dictionary (beyond very basic phonemics), and although they use [ ], the transcriptions are not meant as phonetic (this is clear in the intro to the actual dictionary -- where they also confess to a number of unfortunate compromises they made); check azione, which they give as [a-zió-ne], while the actual pronunciation is [atˈtsjoːne]. WordReference gives phonemic, not phonetic, representations of Italian, such as /atˈtsjone/. You'll continue to be frustrated as long as you refuse to grasp (by now it is refusal, not incapacity) the difference between /atˈtsjone/ and [atˈtsjoːne] (and before you worship the German dictionaries too much, note that PONS gives azione as [atˈtsio:ne], as though it's pronounced with [i], while Langenscheidt ignores that the affricate is long other than for some, definitely not all, northerners: [aˈtsjoːne]; both of those are very unhelpful to the non-native speaker who wants to know real-world pronunciation.47.32.20.133 (talk) 22:43, 1 November 2018 (UTC)

'There's no more need to tell an Italian that the stressed vowel in Roma is long'. No, there isn't. But neither is there any to tell him when it isn't. Nor when it is only 'somewhat' longish. Both being and not being long is allophonic. This I, within limits, do believe (the limits are those of my understanding the very concept of allophony, which is, well, limited, I admit.) Why, then, is the long allophone erected into the 'official' status, in certain dictionaries, and in certain others the non-long one?

Other than the obvious case of phonemic representation (e.g. DiPi and WordReference) vs. (attempted) phonetic representation (PONS and Langenscheidt -- both of which fail in presenting azione accurately), you'll have to take that up with the compilers of the dictionaries. The print version of any dictionary will have a long explanatory introduction; if it's done well, compromises and quirks and warnings will be found therein. Syllabification, for example: do they show orthographic syllabification (e-la-sti-co) or phonological (e-las-ti-co)? Would it be useful to give both, or confusing? Etc. It may be that PONS and Langenscheidt explain what they're up to in presenting the phonetic inaccuracy of [atˈtsio:ne] and [aˈtsjoːne]. Even if they do, they've done no one any favors, and if they don't, the obvious lesson is self-evident: a dictionary is not necessarily a trustworthy guide to pronunciation. Unfortunate, but true. Real example from a dictionary just pulled off my shelf: azione [atˈtsjone]. Why not [oː]? No idea. Vowel length is not mentioned in the introductory matter, and there's no indication that the transcription is phonemic in spite of the use of phonetic brackets. -- BTW, I know absolutely nothing about Polish (other than that there have to be allophonic alternations). Go to the Wikipedia page for Polish phonology, et voilà, examples you can relate to, such as "/x/ has a voiced allophone [ɣ] that occurs whenever /x/ is followed by a voiced obstruent (even across a word boundary), in accordance with the rules given under § Voicing and devoicing below. For example, dach ('roof') is [ˈdax], but dach domu ('roof of the house') is [daɣ ˈdɔmu]." 47.32.20.133 (talk) 15:25, 2 November 2018 (UTC)

Polish has regressive assimilation of obstruents, generally --- so it's not true, literally, that we have no consonant assimilations whatever, yes, even across word boundaries, and 'dach domu' will be automatically realised as [daɣ 'dɔmu], even though most Poles (but some old-fashioned ones, e.g. me) are not capable of pronouncing [ɣ] on purpose and in separation. Yet every Pole always says [daɣ 'dɔmu] for 'dach domu', although, asked what the digraph 'ch' stands for, he'll answer [x]. Is that not called 'complementary allophony' or something like that? A more 'funny' (excuse the vulgarism, please) example is that of the letter 'ń'. It stands -- every Pole will tell you -- for the same sound as does the Italian 'gn', i.e. for [ɲ], yet before a [s] (spelt 's'), such as for instance in the name of my native city of Gdańsk, it always stands for a nasal [j], which most Poles supply automatically but are not aware of having in their phonetic repertoire. So the English WP is wrong with its 'Gdańsk (/ɡəˈdɑːnsk, ɡəˈdænsk/;[1] Polish: [ɡdaɲsk]' as is the German 'Danzig (polnisch Gdańsk [ɡdaɲsk]' and many others, while the French WP is right: ' [ɡdaj̃sk]' Écouter'. (Actually, '[gdaɲsk]' does, if marginally, exist: some clerics have it while officiating, it's probably a kind of a quaint 'spelling pronunciation'.) Now the reason why I am happy with [dax] as the 'canonical' rendering of the pronunciation of 'dach' is that it is much more frequent than [daɣ] and also (I suspect) 'unmarked' (in the sense of the Prague school, if I am not mistaken), again, as distinct from [daɣ], which is marked, viz. by its position directly before a voiced obstruent. And [ɡdaj̃sk] is the ONLY (apart from the 'clerical pronunciation' just mentioned) possible variant of /gdaɲsk/. But revenons à nos moutons: Is ['ro:ma] more frequent than ['roma] and/or unmarked? I am not so sure about it. I personally seem to hear the non-final stressed open-syllable vowels, in Italian that is, significantly less often lengthened than not, see my examples with Collodi and Dante above. And the variants with (noticeable to me) lengthening are marked --- so I understand this term, which you'd probably say I don't -- by their position: isolation, before an intonational pause, or by the pragmatics of their context of occurrence, like in your 'Amerigo Vespucci' example or similar.... Aren't I simply hard of hearing? Maybe. Finally lemme state my ever-recurring 'praeter ea censeo': I don't like the with-lengthening transcriptions because they misguide to mocking and ridiculing one of the most beautiful languages of Europe. 2003:E6:3DA:D921:E567:1390:5A85:59F6 (talk) 13:16, 3 November 2018 (UTC) Wojciech Żełaniec

It would be misguidance to represent both Italian and Spanish 'house' as [ˈkasa] (or Spanish [ˈkasa], Italian [ˈkaza], if you prefer; the point is obviously vowel length).

Yes, it would -- if such transcriptions were to be taken to cover every allophonic peculiarity of both words, and hence, in this case, to imply that both words sound exactly the same. To my ear, be it as dumb, numb, dull, untutored etc. as can be, the Spanish word sounds quite generally different from the Italian one, even if in the latter the /s/ is voiceless (like in the South), and the first [a] is not noticeably lengthened. How come? Well, there are obvious differences, such as that the Sp. [s] has a slight sh-quality to it (you'll know the term-of-art to it) while the It. one hasn't. But that isn't all. The Spanish [a] seems to be somewhat, a wee bit, backer than the Italian [a], though I am not sure if t his holds for 'casa'; and in general, the Spanish vowels sound a bit 'clipped' (a bit like those of British English), which generates the impression (n me) of Spanish being spoken 'more energetically' than Italian, despite its (i.e. of Spanish) approximants (?) spelt VbV, VdV and Vg. Most importantly, the word-stress (a mixture of loudness or 'energy', pitch and length) is realised differently in Sp. and Italian. And of course in a slightly different way from utterance to utterance, speaker to speaker, etc. but I am comparing the general 'strategies' or 'profiles' of word-stress in both languages. And since length in the Italian word-stressing strategy is surely more important than in the Spanish one, you can say that the point is the vowel length, especially when both 'casa's are pronounced in isolation. But I begin to think that in most contexts (technical papers in linguistics aside) when they claim that the only point is the vowel length what they mean is the phonemic length. The difference between 'all' and 'Aal' (eel) in German -- obviously vowel length. Or between German 'gut' (good) and Luxembourgish 'gutt' (good) -- obviously vowel length, quite right. The phonemicity is 'sottinteso', assumed tacitly and taken for granted in such contexts (as it most often is, I surmise) doesn't square up well with Italian and generates confusion in many a poor benighted punter like myself. 2003:E6:3DA:D921:E567:1390:5A85:59F6 (talk) 21:06, 3 November 2018 (UTC) Wojciech Żełaniec

This, almost at the outset of all this, was clearly false: No I am not confusing those two things, I am well-aware of the difference (phonemic/phonetic). You were right immediately below, i.e. that (among many other things) you simply do not -- yet -- understand the concept of phoneme. Why, I have no idea, but resist understanding seems accurate.

I think I sort of misexpressed my ideas (if any...) in English, and this misexpression invited the above comment of yours, but let's leave the matter at that. 2003:E6:3DA:D964:5003:416B:3DB5:90EE (talk) 20:22, 4 November 2018 (UTC) Wojciech Żełaniec

47.32.20.133 (talk) 22:44, 3 November 2018 (UTC)

This I neither refuse nor are incapable of understanding, but simply do not -- yet -- understand, my folksy 'theories' mentioned in this thread put aside.

As for German dictionaries: I do not worship, but rather dread them, yet I have to do with them, for various strange reasons, more often than with the anglophone or polonophone ones.

2003:E6:3DA:D989:3431:7AD0:2E33:4218 (talk) 13:06, 2 November 2018 (UTC) Wojciech Żełaniec

but in all German dictionaries like PONS /quoted already by myself/ or Langenscheidt /e.g. https://de.langenscheidt.com/italienisch-deutsch/dizionario/ the length of the vowel in a non-final open syllable is duly, diliy and conscientiously recognised... In fact I have the impression (accumulated over years and years and years) that the germanophones are more 'rabid' in this respect than the Anglos. 2003:E6:3DA:D928:E816:C28C:19C3:9614 (talk) Wojciech Żełaniec

So, '['ro:ma]' and '[tʰəˈrɑˑntʰoʊ]' are both, in a sense, 'official' and word-in-isolation pronunciations of the city names, although I am not sure if '[tʰəˈrɑˑntʰoʊ]' always and only 'occur[s] at the end of the intonational phrase (thus including isolated words) or under emphasis', as does '[ro:ma]' according to Loporcaro and Bertinetto.

That is not what they said. 47.32.20.133 (talk) 17:52, 1 November 2018 (UTC)

I beg your pardon, sir, but they did say the following: 'The exact phonetic implementation of the stress-conditioned lengthening process is, in any case, prosodically governed even at the word level (Marotta 1985). Thus, although we mark lengthening in our transcriptions, even within stretches of connected speech, this merely indicates the pronunciation appropriate to isolated words' and earlier on they said 'Stressed vowels are lengthened in word-internal open syllables when they occur at the end of the intonational phrase (thus including isolated words) or under emphasis', whence I concluded (fallaciously?) that they would (even if they did not explicitly) say that e.g. '['ro:ma]' is adequate chiefly for 'Roma' in isolation or, more generally, at the end of the intonational phrase, or else under emphasis /all three of which I am consonant with, if a play upon words be here allowed, as can be seen from what I wrote at the beginning of this thread/

Anyway, whoever always said '[tʰəˈrɑˑntʰoʊ]' or '['ro:ma], would be speaking a funny N.American English/Italian.

Of course. yet again: variation is normal. 47.32.20.133 (talk) 17:52, 1 November 2018 (UTC)

So far so good. But you then say that the leap from '[tʰəˈrɑˑntʰoʊ]' or '[ro:ma]' to '['trɑˑnoʊ]' resp. '['roma]' can be achieved with 'just a bit of reflection on variation [...] in his/her own language', and this I doubt. How are we supposed, on the sole basis of reflection like that, to guess that the Canadian city is not be pronounced ['trodo] or some such?

Read what I said. I said nothing about a leap to a specific form such as ['trɑˑnoʊ]. 47.32.20.133 (talk)

No, you did not, sorry. You only mention the general suspicion that the 'canonical', words-in-isolation transcription of a word is -- could not be -- the whole truth. However, I am interested ultimately in avoiding sounding funny in Italian, or rather like mocking Italian -- what way do your own language's variation, diatopic etc., and reflection thereon, help you with that?

Helps in realizing consciously the general principle ~that everyone "knows" in the sense that they do it in their speech and hear it every day: variation is normal. 47.32.20.133 (talk)

in this sense yes, it does help. But not much farther. Especially: it doesn't help much when it comes to tracing out the limits of what is 'really' ('pologise for the vulgarism) normal and what is slovenly speech, slurred speech, drunken-sailor or simply very colloquial, familiar, simplified, careless speech. '[ˈãmõˈtʰɛˑjˑə]' is, meseemeth, off-limits for 'I am going to tell you' as is 'jeet jet' for 'did you eat yet' (I'd go for 'have you eaten yet?'). 'Trahno' is still within-limits or 'really still normal', I guess, but ah dunno [sic]. Now where is '['roma]? Within or off? I'd say 'within'... 2003:E6:3DA:D928:E816:C28C:19C3:9614 (talk) 10:40, 2 November 2018 (UTC)Wojciech Żełaniec

[ˈãmõˈtʰɛˑjˑə] is, meseemeth, off-limits for 'I am going to tell you' as is 'jeet jet' for 'did you eat yet' Yet in large parts of the U.S. both are quite normal in the appropriate context of situation http://www.oxfordreference.com/view/10.1093/oi/authority.20110803095634862 (once jet is adjusted to chet -- no voicing there). Enjoy: https://www.youtube.com/watch?v=wboxzj3TzdA 47.32.20.133 (talk) 20:52, 2 November 2018 (UTC)

' both are quite normal in the appropriate context of situation '. No doubt. There are contexts of situations in which a 'drunken-sailor' or 'potatoes-in-the-mouth' pronunciations are quite, nay, even exclusively appriopriate. Very true! Yet -- I feel that '[roma]' or '[mano]' are legitimate and oftentimes even exclusively legitimate also in quite 'God-fearing and law-abiding', if I may say so, types of situation contexts. 2003:E6:3DA:D921:E567:1390:5A85:59F6 (talk) 13:33, 3 November 2018 (UTC) Wojciech Żełaniec

If you are a (-n a little dull, like myself, but) conscientious language learner, you look up 'mano' in your Langenscheidt, find 'Mano ['ma:no], sb. f. pl. mani, Hand' or some such, take this to heart and ... start saying ['ma:no] whenever ask and unasked-for. Which makes you sound funny. Suppose a friend of yours tells you 'boy, use your hat not just as a hat-rack, reflect a bit on the variations that your native sounds suffer in various parts, style and types of discourse, for goodness' sake' -- will that help you? If you are a Germanophone -- hardly, 'coz in that idiom long vowels stay long. I personally avoided that fate, learning Italian from Polish books where there was no mention of long vowels in Italian, just of word-stress; and that Italians sometimes lengthen their stressed vowels I noticed on my own very soon and learned how to employ /even in Polish, sometimes/; well, is this learning sequence not more practical? 2003:E6:3DA:D928:E816:C28C:19C3:9614 (talk) 20:40, 1 November 2018 (UTC) Wojciech Żełaniec

A polonophone person is particularly at a disadvantage, because Polish has no vowel-quantity, nor vowel-reduction in non-stressed syllables, nor elision (synalefe) not consonant assimilations worth mentioning... and, as I said earlier on, in most languages with non-controversial vowel-quantity, if a word is phonetically transcribed with a ':', then the vowel in question is really long (in the relative sense which I explained above) and stays so, no matter what the context is like. In Italian, by contrast, this doesn't seem to be the case (which is what I take the above sentence by L & B to mean: 'Contrary to wide-spread opinion, this lengthening process is thus far from being a categorical word-level phenomenon'.

What they mean is "do not expect every phonetic realization of Roma to be exactly [ˈro:ma] -- in any case, but especially given that Italian has no phonological vowel quantity" (polite scholarly discourse precludes things like "Good grief, people! Use you heads for something other than hatracks!"). 47.32.20.133 (talk)

2003:E6:3DA:D928:E816:C28C:19C3:9614 (talk) Wojciech Żełaniec

Does 'Italian has no phonological [=phonemic, here? yes] vowel quantity' entail that whoever always says '['roma]' instead of (sometimes, e.g. in isolation) '['ro:ma] -- is bound to sound foreign in Italian ears, may her [r], [o], [m] and [a] be as native-Italian as can be? 2003:E6:3DA:D928:E816:C28C:19C3:9614 (talk) 20:40, 1 November 2018 (UTC) Wojciech Żełaniec

Yes. 47.32.20.133 (talk)

OK. I've got different observations. 2003:E6:3DA:D928:E816:C28C:19C3:9614 (talk) 09:48, 2 November 2018 (UTC) Wojciech Żełaniec

'(Occasionally, however, we use a raised dot to'mark different degrees of lengthening, either to suggest tendential destressing in vowels, or because of performance idiosyncrasies affecting geminates, as witnessed by the recordings.) As for MI (as well as, in general, all regional accents from Northern Italy), it is characterised by tendential lack of vowel lengthening in proparoxytones (cf. tavolo ["tavolo] ‘table’.

Oh yes, again idiosyncrasies and things tendential (but the authors are probably right and use these 'woolly' words legitimately). Hardly ever to be found in descriptions of languages with real vowel quantity.

: Two things. As the article here states quite clearly, In Italian there is no phonemic distinction between long and short vowels. I've reminded you of that several times, and B & L say the same thing succinctly: Italian has no phonological vowel quantity. From your

May I gently remind you, sir, that it's not me that is the primary subject-matter of this exchange? 2003:E6:3DA:D928:E816:C28C:19C3:9614 (talk) 09:48, 2 November 2018 (UTC) Wojciech Żełaniec

first expression of puzzlement about Italian vowel length in comparison to that of some other languages (most notably Finnish, with some very basic phonological characteristics similar to Italian, but with vowel length vigorously phonemic: tuli 'fire', tuuli 'wind'), I suspected you hadn't fully grasped what phonemic (phonologically distinctive) vs. "merely" phonetic meant and implied. Characterizations such as real vowel quantity suggest that you may have a sort of intuitive though muddled grasp of the concept, but that you haven't broken through to overt articulable understanding.

They do suggest that, indeed, ... to the suggestible. Now, certainly, 'real' in the present context is a folksy term, as I (tongue-in-cheek, sordda) used it here. It's got no business jarring in a truly scholarly paper, true enough (but neither have, with your leave, sir, repeated, sometimes jocular, at other times outright confrontational, allusions to one's discussion partner's alleged stupidity, grumpiness or other flaws of character). However, we are not in a scholarly paper here, are we, 'only' in a 'talk' to a WP article, which, in its turn, is not even allowed to be a scholarly paper: the 'original research' ban! And, more importantly: in science (should I say 'real science'?) it's next to always possible to link folksy terms with technical terms, e.g. 'speed' with 'velocity', 'weight' with 'mass', 'wallop' with 'momentum' and so on; not necessarily to 'redeem' or 'gentle' the former by means of the latter, more often to show how confused the former can be and to disabuse their users accordingly. So in 'real' science it's not like: 'But after all, Teacher, how does your formula (thanks be to Goodness for whoever invented it) relate to that thing over there, in the left corner of this class-room? yes, that'un, moving...?' 'Why, apply thee to thy studies for another fourteen years, impudent boy, and thou wilt understand, pupil!' A person, even an excellent scholar, just brandishing her sword of the phonemic-phonetic distinction on high and refusing to hit and cut the knot reminds me of a math teacher from Robert Musil's 'The Confusions of Young Törless' who told one his pupils something like this: 'You may go on studying imaginary numbers for no matter how long, you'll never understand them'. Need I add: obviously he, the teacher, did not understand them himself? Now you may say 'But I have cut the knot!' Well yes, indirectly, viz. by bringing it home to this dumbass, by more rhetorical than scholarly means, that the language game known as 'Italian (anglophone) phonology' has the rule 'long vowels, albeit non-phonemic, exist' as one of its rock-bottom axioms, its mainstays, its 'quos ultra citraque nequit consistere rectum's, in other words, as a premise without whose acceptance and recognition the game can nowise be meaningfully played. 2003:E6:3DA:D928:E816:C28C:19C3:9614 (talk) 09:48, 2 November 2018 (UTC) Wojciech Żełaniec

Quite a screed. And you're not wrong. The glitch in the present case is that there are two levels, or two different types, of reality. If I'm speaking North American English and pronounce tutor so that it sounds like tudor, the phonetic realization of the second "t" is real. If I use the exact same morpheme in tutorial, the second "t" does not sound like the "d" of tudor. What's going on? Allophony. Real phonetic realizations of real phonemes. Go to Tuscany and you'll hear [k]asa, la [h]asa, tre[kk]ase -- three phonetic realizations of the same phoneme, all real. Basta! 47.32.20.133 (talk) 15:52, 2 November 2018 (UTC)

OK. I have heard the Tuscans with their 'hasas', liked them, reminds me, purely phonetically, of the dreaded Celtic mutations. Well, we have reached a certain rapprochement, haven't we?, in that neither of us denies what the other puts in the foreground /I don't deny that Italians quite often lengthen their stressed vowels in non-final open syllables and you subsume their shortening them as admissible allophonic variation/, the difference being still that which you resp. I do put in the foreground. Since chances of a 100 p.c. consensus don't seem very high within foreseeable time -- it's time to say basta and thank you! I have learnt from you a lot. 2003:E6:3DA:D921:E567:1390:5A85:59F6 (talk) 15:08, 3 November 2018 (UTC) Wojciech Żełaniec p.s. this is a bit tangential but you certainly know that some Brits indulge in mocking American speech by drawling and diphthongising their vowels, like 'Bawahston' for 'Boston or 'laierm' for 'lamb'. Yet Americans do drawl their vowels, don't they? So how is the 'mocking' possible? My dream is to find a description of the Italian vocalism which does justice to the intuitions of those who /like you/ believe in vowel quantity in Italian, and then does not encourage mockery. 2003:E6:3DA:D921:E567:1390:5A85:59F6 (talk) 15:08, 3 November 2018 (UTC) Wojciech Żełaniec

No one has control over clowns who choose to mock. And I don't "believe in" Italian (or other) phonetic facts. I observe them. 47.32.20.133 (talk) 16:23, 3 November 2018 (UTC)

Good for you. But, at the risk of its being me this time that is going to sound like Mr. Know-It-All, theory of science being somewhat closer to my profession than is linguistics: You certainly know that pure, theory-free and not theory-primed observation is rather difficult to come by. But OK, you have heard many phonetically long or very long vowels in Italian, so have I, but even if those were 'naked', non-theory-laden observations, it is not THEM that we have been discussing in this exchange, but, rather, various bits and pieces of a theory that was constructed on their basis, and which as such (as a theory) goes beyond anything that either of us has ever observed, and may these observations, especially yours, have been as pure and theory-free as one can possibly wish. Gosh, evidently a topic difficult to take one's mind off,since we've said so often 'basta' yet go on... 2003:E6:3DA:D921:E567:1390:5A85:59F6 (talk) 21:32, 3 November 2018 (UTC) Wojciech Żełaniec

various bits and pieces of a theory that was constructed on their basis, and which as such (as a theory) goes beyond anything that either of us has ever observed. No, you speak for yourself only. I've been describing from decades of observation, not theorizing. 47.32.20.133 (talk) 22:44, 3 November 2018 (UTC)

OK. Speaking for myself (tho' I hate egotism of all kinds): I believe in the existence of the vowel /a/, phonetically [a], or even [ä] in many languages (not sure about English), though I never met it 'in person', i.e. I never observed it. What I did observe so far were various individual 'sounds' (to speak with the vulgar) that most linguist would call 'individual instances' or 'tokens' (in C. S. Peirce's terminology) of /a/ or even [a]. But /a/ or [a] is an abstract entity, a Platonic idea or an Idealtypus, to speak with M. [ˈveːbɐ], or some such, in any event nothing accessible to the senses alone, but arrived at as a result of some theorising. That's why I prefer to call it a theoretical construct. I don't construct such constructs myself, needless to say, so I seem to be just observing whatever they are relevant for, but I more or less competently avail myself of such constructs, and if I have learnt them 'down pat' this comes naturally and I subsume the results of my observations under such constructs 'on the fly', fluently, and so I am able to immediately 'see' (hear) that the CAT-vowel occurs thrice in 'Jack Spratt could eat no fat' as just uttered by my nephew Władek. But the CAT-vowel, whether as a phoneme, or in an ever so 'narrow' phonetic version, is not observable as such. It is a construct in the sense here (certainly rather clumsily) set forth. Whoever constructed this construct must have been a genius, which I am not, and far from that (and that's why I am discouraged from thinking up new theories and encouraged to just use old ones, such as.... e.g. that there is, in Italian, Polish, Northern British English ... (you name them) the vowel /a/, which no-one ever heard or ever will.). Does this make any sense to you? 2003:E6:3DA:D964:5003:416B:3DB5:90EE (talk) 20:22, 4 November 2018 (UTC) Wojciech Żełaniec

Until you do, the variation licensed by the non-distinctive character of Italian vowel length will most likely continue to bother you.

OK, I understand THAT, roughly, of course. The length and shortness of vowels in Italian is allophonic, isn't it? Fine, but so are many-many-many other phon(et)ic traits of the sounds Italians produce while speaking. Why not put them all into the '[...]' (phonetic) transcription of Italian entries in a dictionary. Or -- given that that's sheerly impossible -- why not put many more of them into it? Why select the non-phonemic lengths (and as I noticed before, that is done not uniformly, by every scientifically responsible publisher on this planet, but only by some of them, mostly from countries whose official languages do have phonemic vowel-quantity) amongst many other phonetic but not phonemic traits of Italian sounds, in publications directed to non-just-technically-linguist audience? Now L & P do give some reasons (position in the phrases, emphasis, etc., I with my untutored ear have heard the same as them, long since) and you, too, have mentioned others, to wit: pragmatic factors, such as the one below with 'Amerigo Vespucci'. Great. But this shows that this issue is not a non-issue and if it isn't, then why on earth should a (civil) discussion about it be considered 'polluting' and 'whaffling'? 2003:E6:3DA:D928:E816:C28C:19C3:9614 (talk) 09:48, 2 November 2018 (UTC) Wojciech Żełaniec

Second, B & L reminding you of what you have observed and I confirmed to you, that vowel length is often absent in the speech of many northerners is a totally unsurprising result of the non-phonemic status of Italian vowel length (not to mention the diatopic variation found in any language).

Funny you should mention it, but I have often had the impression that in certain forms of popular 'lumbaart' speech (half-dialect?) there IS a phonemic vowel length, probably due to the disappearance of phonemic consonant gemination, so 'Fano' and 'fanno' would sound the same were it not for 'Fano's long [a]. But I have too little observation for that. Buona giornata from the Valley of the Benighted to the Plateau of the Presumptuous, sir! 2003:E6:3DA:D928:E816:C28C:19C3:9614 (talk) 09:48, 2 November 2018 (UTC) Wojciech Żełaniec

-- Ora basta, per favore. It's way past time to stop polluting this talk page with this repetitive waffling. If there's anything you want to discuss further -- discuss seriously, with the goal of understanding -- just take it to my talk page. 47.32.20.133 (talk) 17:59, 1 November 2018 (UTC)

Thank you for your generosity and magnaminity, sir, but I shall not. 2003:E6:3DA:D928:E816:C28C:19C3:9614 (talk) 09:48, 2 November 2018 (UTC) Wojciech Żełaniec

Por fin! Someone should work in this excerpt into the article, at the relevant place (viz. where the author says that stressed non-final vowels are long, full stop.). And then this whole overlong thread can happily be deleted from the 'Talk'. ALL'S WELL THAT ENDETH WELL and 'schlafe, was willst du denn mehr?' 2A02:A312:C43D:DC00:D92C:8EB4:9368:3B5D (talk) Wojciech Żełaniec —Preceding undated comment added 23:17, 31 October 2018 (UTC)

Or it's like indicating word-final devoicing in German and Russian. Or final stress in French. Or omitting linking r in non-Rhotic English. Or word-final rhotics as [r] when they often become [ɾ] when a following word begins with a vowel. Dictionaries often have to make compromises like that and there's typically what's called a "citation form" (see lemma (morphology). A really extreme example is the "incomplete" and "complete" forms of the Rotuman language. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 16:21, 1 November 2018 (UTC)

My main point was (and continues being) this: The pronunciation of the Italian citation form is (thinkz I) preferably one which does not contain the ':' sign, because: 1. The phonetic length of the vowels of non-final open stressed syllables in Italian is (pardon the vulgar way of putting it, I am not a professional linguist) exclusively a form in which the word-stress of the stressed vowel manifests itself, and it can manifest itself in different ways in non-isolated or non-final positions occupied by the same word. This is not true for most European languages in which there is non-controversial vowel-quantity, although in all (afaik, of course) of these languages the being longer of the long vowel is always relative (to/on the absolute length of the adjacent vowels) and cannot be measured in milliseconds. 2. buyers of dictionaries with transcriptions like '['ma:no] for the pronunciations of the citation or dictionary-entry form of 'mano' will -- certainly not all, but if my anecdotal evidence from many years' observation is not totally untrustworthy, quite a few -- be misinstructed to saying consistently ['ma:no], regardless of whether the given 'mano's position is isolated or phrase-final or meant-as-an answer-to-the-question-'Who gave America its name?' or some such. And this will give their Italian a funny (sorry for the vulgarism once again) sound and even make them appear as if they wanted to mock Italian or the Italians. So I think that's a sub-optimal compromise, alas quite frequent in germanophone and to a perhaps somewhat lesser extent in the anglophone countries. It's better, I suggested, to write simply ['mano], explaining to the reader in the preface that word-stress in Italian is realised in such-and-such fashion (which she will pick up on herself quite soon, anyway). And I must say that I hate hearing Italian(s) mocked, unless, perhaps, the mocker is an Italian himself, like Toto Cutugno (https://www.youtube.com/watch?v=syc78JzHGTs), 2003:E6:3DA:D928:E816:C28C:19C3:9614 (talk) 08:24, 2 November 2018 (UTC) Wojciech Żełaniec

Georg Christoph Lichtenberg coined a word for acting as editors do here: verschlimmbessern (Grimm – Duden – en.wikt – pl.wikt). Love —LiliCharlie (talk) 10:32, 2 November 2018 (UTC)

Yes. Except that from my point of view (and I think not just mine) what is the Verschlimmbesserung /disimprovement?/ is the very introduction of the colon in the Italian transcription. A lesser evil would, I think, be to take out the stress mark and leave the colon, as marking word stress in this type of syllables by itself. 2003:E6:3DA:D989:3431:7AD0:2E33:4218 (talk) 15:10, 2 November 2018 (UTC)Wojciech Żełaniec

What they do here is like writing ¹H₂¹⁶O when they mean H₂O. Though it is true that 99.98% of hydrogen isotopes are ¹H and 99.76% of oxygen isotopes are ¹⁶O it is untrue to say that ¹H₂¹⁶O is the formulaic representation of dihydrogen monoxide; the supposedly "more exact" formula covers only about 99.74% of dihydrogen monoxide molecules, and a notation that covers exactly 100% of them is available. Love —LiliCharlie (talk) 03:23, 3 November 2018 (UTC)

Yes; well, ['ro:ma] or even ['ro::ma] are probably more frequent than 0.26 p.c. of all ocurrences of 'Roma' and 'mano', in all fairness... 2003:E6:3DA:D921:E567:1390:5A85:59F6 (talk) 13:33, 3 November 2018 (UTC) Wojciech Żełaniec

This practice leads to transcribing long vowels even in syllables that are clearly not at the end of the intonational phrase, e.g. [ameˈriːɡo veˈsputtʃi] in the Amerigo Vespucci article, which violates Bertinetto & Loporcaro (2005). It is easy to find tons more of those. Love —LiliCharlie (talk) 23:49, 1 November 2018 (UTC)

That's reading a bit too much into what B&L say (and it's on them; they could have been clearer). In citation mode (answer to "What's that guy's name?"), the length in [ameˈriːɡo vesˈputtʃi] is normal. 47.32.20.133 (talk) 01:05, 2 November 2018 (UTC)

Off-topic P.S.: [veˈsputtʃi] is also strange; [vesˈputtʃi] would seem more reasonable, cf. the last paragraph of the Italian phonology#Onset section.

Yes. It's more than a bit thorny, but forced to come up with either .sC or s.C as syllabification of s+C clusters word-internally, s.C is the only justifiable choice -- Cas.tel.lo, etc. There's a crank who'll chase you around and revert if you try to change these (he cites orthographic conventions for hyphenation rather than phonology, without understanding the difference), one amusing bit being the case of Pescara, where he insists on Pe.skara, even though in the audio right next to it, it's very clearly Pes.kara. 47.32.20.133 (talk) —Preceding undated comment added 00:27, 2 November 2018 (UTC)

Stress and phonotactics

This article is missing information on stress (or timing), as well as phonotactics of Italian and such information should be added. EditWorker (talk) 12:00, 28 November 2018 (UTC)

Yeah, go ahead. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 16:50, 28 November 2018 (UTC)

I agree, in spoken Italian there is the distinction between stressed vowels (à, á, è, é, ì, ò, ó, ù) and unstressed vowels (a, e, i, î, o, u) (in written italian only a, à, e, è, é, i, ì, î, o, ò, u, ù).
The distinctions between è, e and é is the distinction between ((/e~e̞~ɛ/) and the distinction between ò, o and ó is the distinction between (/o~o̞~ɔ/); the letter î (same as the letter y) represents the /j/ approximant palatal phoneme, in the page is completely missing the distinction between à, a and á (/a~ä~æ~ʌ/), between i and ì (/i/) and between u and ù (/u/), the better way is to add an entire section only for stress distinction.
(the IPA distinction between vowel pronunciations is not sufficient to accurately express the distinction between Italian pronunciations, especially for the distinction between ì, i and í and the distinction between u and ù)
Stresses that are not at the end of a word are written only in vocabularies for the hyphenation and for distinction between omographic terms and in non-standard writing; the vowels á and ó are only in vocabularies also when they are at the end of a world, the vowel î is rare outside handwritten texts because in the italian standard keyboard it is missing although it is used in the plural of some words principi is the plural of principe while principî is the plural of principio, similar distinction is between tempi plural of tempo and tempî plural of tempio (in modern texts can be found templi as alternate form of tempî).
The difference between stressed and unstressed "i" and "u" may be a tonal distinction. --DelvecchioSimone12 5 96 (talk) 22:07, 6 February 2019 (UTC)

Alveolo-palatal or palato-alveolar?

The heading of the consonant table links to Alveolo-palatal consonant, but the symbols used are for Palato-alveolar consonants. The rest of the text doesn't mention either term. So which is it? --Ørjan (talk) 02:13, 16 March 2019 (UTC)

Oh yeah, we should be clear about that. I believe the nasal and lateral are alveolo-palatal, the approximant is true palatal, and the affricates and fricative are palato-alveolar. Should we just go back to linking to only palatal in the table? — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 17:03, 16 March 2019 (UTC)

Thanks, after Nardog's edit it no longer looks misleading, although the text doesn't mention your more precise descriptions. --Ørjan (talk) 13:32, 21 March 2019 (UTC)

Ahh, true. Okay, I've added the note with a citation. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 15:41, 21 March 2019 (UTC)

A coda is only permissible in case of monophthong nuclei

In light of e.g. fuorché, this could use some elaboration. --47.32.20.133 (talk) 20:35, 31 July 2018 (UTC)

As of today as I write, the Coda section is even more problematic. The ban on diphthongs (resulting from diphthongs having arisen historically only in open syllables) was perhaps somewhat otiose, but pretty much true. The text now reads A coda can be one of: and supplies an incomplete list. Barefoot through the chollas (talk) 22:52, 28 April 2019 (UTC)

Following at our description of onsets ending in /j/ and /w/, /jV/ and /wV/ are not diphthongal nuclei, but sequences of an approximant consonant belonging to the onset plus a (monophthongal) vowel that constitutes the nucleus. Love —LiliCharlie (talk) 23:20, 28 April 2019 (UTC)

Ignoring the circularity of declaring that [jV] and [wV] are not diphthongs, ergo..., the diphthong controversy is peripheral to the main point of the incomplete list of codas. More than a tiny bit of jiggery-pokery is necessary to limit codas to just /r/, /l/, /n/, /m/ and first constituent of geminates (even more jiggery-pokery than is necessary to demote all [jV] and [wV] out of diphthong status). Barefoot through the chollas (talk) 02:52, 29 April 2019 (UTC)

I've added some clarifying content. Kramer (2009) sees the fact that [jV] and [wV] do not pattern with [Vj] and [Vw] in coda restrictions indicates that the former pair are ones of complex onsets, rather than diphthongs. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 03:46, 29 April 2019 (UTC)

Better, thanks. I'm pressed for time just now, but I can mention a few things that still need work, with two matters in mind: simplicity (comprehensible to readers not steeped in Italian phonology?), accuracy. Perhaps a list of items that need to be covered in the descriptions will help: asfalto, risveglio, nafta, optare, maybe ictus if a realization with [kt] is not to be kicked into the learnedism bin on grounds that [ˈittus] is common. What to do with [ˈspiando] ~ [ˈspjando] (in light of /ˈspi.a/); treating the type as phonemically distinct, as quieto /ˈkwjɛto, kwiˈɛto/ under CCC is... [I'm having trouble thinking of a polite euphemism], and it's very easy to get [i] in proprio. The bit on nasals can be simplified to just /n/ and /m/, as those two cover all phonetic realizations. Weaved in with a lot of this is treatment of the sibilant of sdraiare etc. as /z/, yet this clear statement earlier in the article: The phonetic distinction between [s] and [z] is neutralized before consonants and at the beginning of words: the former is used before voiceless consonants and before vowels at the beginning of words; the latter is used before voiced consonants (meaning [z] is an allophone of /s/ before voiced consonants); consistency in recognizing the standard /s/ → [z] assimilation would greatly simplify various treatments under Phonotactics (and account for reality). I'm tempted to nitpick the entire section CCC into (near) non-existence, but I do have to run. Barefoot through the chollas (talk) 14:51, 29 April 2019 (UTC)