This page is an
archive of past discussions for the period 25 December 2017 - 21 January 2018. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the
current talk page.
Alveolo-palatals
While I agree they don't deserve a column in the IPA consonants table, simply because the term is limited to fricatives and affricates so the column needlessly widened the already crowded table, they definitely have to be listed somewhere on one of the tables. Should they be moved back to the co-articulated consonants? Or just outside the pulmonics/affricates tables?
Nardog (
talk)
17:01, 25 December 2017 (UTC)
@
Nardog: It's not limited to fricatives and affricates. All consonants marked as palatal on the official IPA chart can be alveolo-palatal (apart from [cç,ɟʝ,ç,ʝ] and perhaps [j̊,j], but that may be debatable). [tɕ,dʑ,ɕ,ʑ] can stay where they are right now, because they're sibilant counterparts of [cç,ɟʝ,ç,ʝ]. The classification is correct.
Mr KEBAB (
talk)
17:19, 25 December 2017 (UTC)
Ah, you're right, my mistake. I also missed the fricatives and affricates were moved to the sibilant rows. I'm fine with the current layout. Sorry to have bothered you.
Nardog (
talk)
17:26, 25 December 2017 (UTC)
That would be consistent with other alveolo-palatals (those which do not have their own IPA symbols). But the thing with [(t)ɕ, (d)ʑ] is that they sound way closer to [(t)ʃ, d(ʒ)] than to [(c)ç, (ɟ)ʝ]. I wonder why only the fricatives are given their own IPA symbols, but that would be my best guess (and perhaps they're more common cross-linguistically?). And I would want to know why before we decide to merge the articles for [(t)ɕ, (d)ʑ] with [(c)ç, (ɟ)ʝ].
Nardog (
talk)
13:34, 26 December 2017 (UTC)
@
Nardog: That depends on what you consider [tʃ,dʒ,ʃ,ʒ] to be. IMO, considering canonical [tʃ,dʒ,ʃ,ʒ] to be weakly palatalized is just weird. It's like saying that canonical [l] is palatalized or that canonical [n] is velarized. The coloring should be neutral. Plus, no language contrasts weakly palatalized [tʃ,dʒ,ʃ,ʒ] with [tɕ,dʑ,ɕ,ʑ], but there are languages (Polish, Mandarin, etc.) that contrast hard, completely unpalatalized [tʃ,dʒ,ʃ,ʒ] with [tɕ,dʑ,ɕ,ʑ].
If you consider canonical [tʃ,dʒ,ʃ,ʒ] to be hard, then [tɕ,dʑ,ɕ,ʑ] are much closer to [cç,ɟʝ,ç,ʝ]. If you don't, they sound as something in between them, but a bit closer to [tʃ,dʒ,ʃ,ʒ] because of their sibilant nature.
Mr KEBAB (
talk)
14:08, 26 December 2017 (UTC)
@
Nardog: On second thought, considering [tʃ,dʒ,ʃ,ʒ] to be weakly palatalized is much weirder because these variants are much closer to [tɕ,dʑ,ɕ,ʑ] than the flat postalveolars. I hope the IPA clarifies this should they ever write the next edition of the Handbook.
Mr KEBAB (
talk)
15:15, 8 January 2018 (UTC)
@
Nardog: I'd put that on hold. We should merge the articles on [tʃ,dʒ,ʃ,ʒ] with the ones on [tʂ,dʐ,ʂ,ʐ] - see
User:Mr_KEBAB/VPAS (it's far from being 100% finished). It's a bit too early to start a discussion about that on
Wikipedia:WikiProject Linguistics, and I don't want time pressure to finish all of the four articles. If editors agree on the merge (there are good reasons for it), then we can think of appropriate article names, which will be a minor issue anyway.
Mr KEBAB (
talk)
18:50, 25 December 2017 (UTC)
As far as other postalveolar consonants are concerned, I'd also move them to the corresponding articles about retroflex consonants. Maybe the articles on [ɹ] and [ɻ] should be fully merged, I'm not sure.
Mr KEBAB (
talk)
18:52, 25 December 2017 (UTC)
I didn't realize you were prepping those articles, sorry if I stepped on your feet. But wouldn't you at least agree there need to be articles named "voiceless/voiced postalveolar fricative/affricate"? Your wording ("merge ... with the ones on [tʂ,dʐ,ʂ,ʐ]", "...move them to the corresponding articles about retroflex consonants") suggests you're intending to merge the postalveolars into articles named "...retroflex...", which seems odd. Don't you actually mean the other way around? Or do you actually think they should be gathered in articles that have "retroflex" in their names?
I also have a feeling it wouldn't be a bad idea to accumulate only truly retroflex (subapical) sounds into the articles on retroflexes, provided that they are common enough to warrant the articles, though I don't know how practical it is to ascertain the actual tongue shape of each sound described as retroflex.
Nardog (
talk)
13:34, 26 December 2017 (UTC)
No wonder you didn't, the only person I told about that was No such user. I only said that because the merge is very likely to occur, so we'll rename them again anyway.
I have no idea how we should call them. I'm counting on other editors to choose the names. Lazy, I know.
Ok, but the very reason I want to merge the articles on postalveolar/retroflex sibilants is that flat postalveolar sibilants found in Polish, Lithuanian etc. are sometimes called retroflex. The corresponding Mandarin sounds are always called retroflex, whereas the corresponding French sounds are never considered to be retroflex. But they're all the same sounds pronounced with flat tongue. (EDIT: That's irrelevant to what you wrote) As far as other retroflex consonants are concerned, that strikes me as severely limiting. What about non-subapical retroflexes/postalveolars that are transcribed with the retroflex symbols?
Mr KEBAB (
talk)
14:08, 26 December 2017 (UTC)
Frankly―and you seem to agree on this―the current organization of speech sounds is an utter mess. You've already touched upon this idea at
WT:LING but there are a lot of scarcely attested sounds that are better merged with related sounds. I think [
ä and [
ɒ̈ can merge too. There are also completely unattested
Labiodental trill etc., which I think should be redirects to an article called
Unattested sound or something to that effect (or to some existing article).
Nardog (
talk)
13:34, 26 December 2017 (UTC)
Yeah. I screwed up by not merging [ɪ̈,ʊ̈] with [ɘ,ɵ] a few months ago when I said I would. But the articles are almost ready and the merge can still be performed. I think that [ɯ̽] and [ɤ] can stay separate. At least I think that's what you're talking about.
I'd merge [ɒ̈] with [ɞ̞] but not necessarily [ä]. [ɶ,ɑ,ɒ] also cover both open and near-open sounds, but it's good to preserve the rounded-unrounded distinction, which is consistent with the rest of the articles.
Mr KEBAB (
talk)
14:08, 26 December 2017 (UTC)
Actually, now that I think about it, I'm not so sure if I would like "merging [ɪ̈,ʊ̈] with [ɘ,ɵ]". As far as English is concerned, [ɪ̈, ʊ̈] are better transcribed as [ɨ̞, ʉ̞] (or even simply [ɨ, ʉ]) because ⟨ɨ, ʉ⟩ aren't occupied by other phonemes. Are [ɪ̈, ʊ̈] known to contrast with [ɨ, ʉ] in any language? If not, I feel like they are better merged with [ɨ, ʉ] than with [ɘ, ɵ].
@
Nardog: The reason [ɪ̈,ʊ̈] should be merged with [ɘ,ɵ] is that [ɪ,ʊ] cover both near-close and close-mid vowels (see
near-close near-front unrounded vowel and
near-close near-back rounded vowel). Because of that, it's pretty much impossible to determine whether vowels transcribed with [ɪ̈,ʊ̈] (or the corresponding non-IPA symbols [ᵻ,ᵿ]) are near-close or close-mid unless that is explicitly stated in a given source. I agree that you should expect[ɪ̈,ʊ̈] to be near-close, but that's not good enough for Wikipedia.
There's also the issue of the centralizing diacritic itself, which some authors (e.g. Collins & Mees in their Phonetics of English and Dutch) use to mean mid-centralized, not centralized, or even both. For example, they say that the pre-/r/ allophone of /i/ is [ïə], but it's actually [ɪə] (at least in Randstad, other speakers probably use a fully close vowel with an optional schwa offglide). So even when we know that a scholar considers [ɪ,ʊ] to be near-close, we can't really be sure that their [ɪ̈,ʊ̈] are near-close as well. So it makes a lot sense to merge [ɪ̈,ʊ̈] with [ɘ,ɵ]. It's consistent with the series of edits I made to [ɪ,ʏ,ʊ] almost a year ago.
When we merge [ɪ̈,ʊ̈] with [ɘ,ɵ], we'll be also able to list more English dialects with central rounded FOOT and NURSE. This is good for our readers. Plus, the variants of FOOT transcribed with [ɵ] and [ʊ̈] may not be different at all (different sources use different notations, very often without providing vowel charts).
Hi, long time no see! I wonder what's your opinion on Polish (and maybe other Slavic) names pronunced in English? For example,
Łódź beling pronunced like
WOOCH? Do they sound strange or foreign? I'm very interested to hear your opinion. – they call meAWESOMEmeeos ...
[ˈɔɪ̯]!
12:09, 29 December 2017 (UTC)
@
Awesomemeeos:[wuːtʃ] for Łódź does sound foreign. It must be so, because only the first sound is correct as far as Standard Polish is concerned. But we're not talking about a non-native pronunciation of a Polish name in Polish, but an English approximation thereof. It's a big difference. I'd have no problem with hearing [ˈbɪdɡɑʃtʃ] or [ˈbɪdɡɑʃ] for Bydgoszcz and [ˌʃvɪnɵʊˈʉʃtʃə] for Świnoujście. Only someone that doesn't know how phonetics/phonology work would. But there's no excuse for not getting the stress right.
Mr KEBAB (
talk)
13:31, 29 December 2017 (UTC)
1) It seems that words in onomastika are proper names, but geographical names like the mountain Konjuh are also proper names... So I am not sure if they are given names and surname...
@
LoveVanPersie: I'm not sure if there's any English word that begins with /ər/ as opposed to /ɜːr/. It seems to me that only the latter is allowed in the unstressed word-initial position. @
Nardog:, can you confirm that? You have a better grasp of phonology than me. The non-rhotic pronunciation in the third link is definitely /ɜːrˈnɛstoʊˌɛskəˈbeɪdoʊ/, with strong /ɜːr/, not weak /ər/ (phonetically [əː] and [ə] in the narrator's accent, but you probably know that already). Bear in mind that in much of the North America there's no actual distinction between /ər/ and /ɜːr/. There's a related English name Ernest by the way, and it's pronounced /ˈɜːrnɪst/, with the same /ɜːr/ but stressed (/ər/ can't be stressed in accents that contrast it with /ɜːr/).
The correct respelling is ur-NES-toh ES-kə-BAYD-oh. Stressed /ɛ/ can never occur in the syllable-final position.
It's initial in both cases, but it seems to me that only the female pronouncer treats the first name as trisyllabic ([ˈsaːnɪaː]). The man seems to compress /-nɪ.aː/ into [-njaː], yielding [ˈsaːnjaː]. This is very similar to what happens to unstressed /i/ before /ə/ in English (see
Help:IPA/English).
1. For surnames, look under pr. I'm not sure what it stands for, but that's where they are. For first names, look under m. os. ime (for male first names) and ž. os. ime (for female first names).
2. They're two different surnames. Hers is Martić with ć.
3. Jȕrak is diminutive of Juraj. The surname is Jȕrāk, with a long post-accentual a.
4-6. Both pronunciations are correct.
7. I'd only transcribe Bȁšić. I don't know about the other variant. It could be that Elly Bašić insisted on pronouncing her surname Bàšić like
Rachel Weisz insists on the /vaɪs/ pronunciation of her surname. Maybe it has dialectal connotations. These are all speculations by the way. Mail HJP if you want a definitive answer.
You have a better grasp of phonology than me. Well, that's flattering but probably not true. Anyways, I think you're right, I can't think of any instance of word-initial /ər/. Searching in CEPD/LPD seems to confirm that. Happy new year.
Nardog (
talk)
06:05, 4 January 2018 (UTC)
LoveVanPersie That's almost correct. The IPA is /θəˈnɑːsiˌkɒkɪˈnɑːkɪs/ (thə-NAH-see KOK-ih-NAH-kis). I've noticed that you're still confused about the lax vowels, which can't ever end stressed syllables (save for extremely rare instances of loanwords such as pho). You don't have to use the IPAc-en and respell templates here by the way. You're making an effort that doesn't make a difference (I don't mean to sound disrespectful, I'm just trying to save your time). The reason we use the IPAc-en template in articles is because of the
mouseover feature and because it links to
Help:IPA/English where all of the symbols are explained. In the case of the respell template, we use it mainly because it links to
Help:Pronunciation respelling key.
Yes, the spelling itself proves it. Catalan doesn't have ⟨ñ⟩, they use ⟨ny⟩ instead (interestingly, that feature is shared with some African languages). Compare Catalan Catalunya[kətəˈɫuɲə] with Spanish Cataluña[kataˈluɲa].
Mr KEBAB (
talk)
23:53, 4 January 2018 (UTC)
The correct syllabification seems to be /ˈkVr.i.ɒs/ (where V stands for vowel), so /ˈkiːriɒs/ is probably impossible. The only possibilities are /ˈkɪriɒs/ and /ˈkɪəriɒs/ (Australian [ˈkiɹ-] and [ˈkiːɹ-], respectively), and I think it's the former. The dialect of English I speak ("I'm trying to imitate" is probably a more suitable phrase) doesn't have phonemic vowel length, so it's a bit hard for me to distinguish Australian /ɪ/ from Australian /ɪə/, as they have pretty much exactly the same quality and differ only in length. Strangely, I don't have such problems with English/ɪ/ and /ɪə/ in the same positions. Apparently I still have more to learn as far as AuE is concerned. /ɪ/-tensing still sounds strange to my ears.
Here, the pronunciation with /ɪr/ is more obvious, as the commentator is British. I'm 95% sure that the correct IPA is /ˈkɪriɒs/, with the lax /ɪ/.
Mr KEBAB (
talk)
01:01, 5 January 2018 (UTC)
@
LoveVanPersie: Nope, unstressed syllable-final /ɒ/ appears only before consonants. The correct IPA is /ˈtrɔɪsi/ (TROY-see). The [ɪ] is definitely non-syllabic ([ɪ̯]) and short.
Mr KEBAB (
talk)
01:26, 5 January 2018 (UTC)
Sorry to nitpick, but /iː.r/ is certainly possible albeit rare, most notably in some Americans' pronunciation of hero and zero (see AoE pp. 481–2). Parallels to this are also found in Nero, pharaoh, and guru.
Nardog (
talk)
02:56, 5 January 2018 (UTC)
Ah, that's what you meant. I agree the name does seem to have MIRROR, at least judging from the UK pronunciation in the video you linked to.
Nardog (
talk)
03:25, 5 January 2018 (UTC)
@
LoveVanPersie: 1. Yes, but only when there's a pause between the words, as on the recording. The correct IPA for the variant without a pause is [ˈnorberdˈɡomboʃ] (notice that ⟨ɡ⟩ is the correct transcription, not ⟨g⟩).
2. Yes.
3. Yes.
4. No, it's /ˈstɛfɑːnˈkɒzlɒv/. What we transcribe /ɑː/ on
Help:IPA/English can never be written with ⟨o⟩.
5. Visit
Help:IPA/Slovak and
Slovak orthography and tell me the IPA. It's straightforward (remember that the surname is written Kovalík, with ⟨í⟩).
6. The guy has a non-native accent and so the recording can't be used as a source.
@
LoveVanPersie: Our
diaphonemic transcription forces us to write /ˈkɒzlɒv/. From it, you can easily deduce that the General American pronunciation is [ˈkɑːzlɑːv].
"Consonant clusters containing both voiced and voiceless elements are entirely voiced if the last consonant is a voiced one, or voiceless if the last consonant is voiceless." But pr and kr seem exceptions?
LoveVanPersie (
talk)
06:02, 11 January 2018 (UTC)
@
LoveVanPersie: I'm not knowledgeable enough to discuss Slovak phonology with you. Try to get your hands on The Lexical Phonology of Slovak which should answer your question. Perhaps
sonority hierarchy will answer that question too.
And should we need to remove the tie bars on
Help:IPA/Mandarin, where [t͡ɕ], [t͡ɕʰ], [ʈ͡ʂ] and [ʈ͡ʂʰ] are with tie bars but [ts] and [tsʰ] are without them. In
Template:IPAc-cmn, transcriptions for the six affricates are all left out tie bars.
Is the /v/ in Viktória pronounced [v] or [ʋ]? I think it's a normal situation so it is pronounced [v]. But it seems a case of "[ʋ] occurs in all other cases". So I'm confused...
Is the /m/ in Škamlová pronounced [m] or [n]? I think it's the former according to
Slovak phonology, in which /m/ only "has the allophone [ɱ] in front of the labiodental fricatives /f/ and /v/".
Is Rebecca Šramková pronounced [ˈrebet͡sa ˈʃraŋkoʋaː]? And is
Šramková a Czech or Slovak surname?
@
LoveVanPersie: The answer is on
Help:IPA/Slovak, which you seem to be quoting. The note you're quoting is unambiguous, so I don't understand why you're saying that it's a normal situation so it is pronounced [v] (whatever you mean by normal situation). There's nothing to be confused about.
/m/never assimilates to [n] or [ŋ] in standard Slovak. It'd be a strongly abnormal pronunciation that doesn't occur in any language I'm aware of.
Please don't ask me about the pronunciation of Slovak words.
Radoslav Ivan is a better candidate for that, but bear in mind that I am somewhat abnormally generous when it comes to answering your questions. Radoslav might not have that much time or patience.
Once again, you're asking me a question you could've answered yourself. See
Šramková and pay attention to the spelling.
Also, are you 100% positive that the first name of
Chantal Škamlová is pronounced with [x], not [ʃ]? It seems to be a French name. The same applies to
Andrej Martin - are you sure that the surname is pronounced with [t] and not [c]? The research we did was on a first name, not a surname. They don't have to be pronounced in the same manner (see
homograph). What I'm saying is that you should never add IPA if you're not sure if it's correct. It could be taken as
WP:VANDALISM and get you banned.
Mr KEBAB (
talk)
19:30, 12 January 2018 (UTC)
@
LoveVanPersie: But you should've expected that. Hardly any language respells every single loanword. The Serbian variety of Serbo-Croatian does, Slovak doesn't. It doesn't even have a fully phonemic orthography.
Mr KEBAB (
talk)
15:58, 14 January 2018 (UTC)
@
LoveVanPersie: The first name is pronounced /bərˈnɑːrdə/, but I'm not sure about the surname. GA doesn't distinguish /ɛr/ from /ɛər/, so it's either /ˈpɛrə/ or /ˈpɛərə/. It'd be perfect if I could hear the British pronunciation of the surname.
Mr KEBAB (
talk)
09:34, 18 January 2018 (UTC)
@
LoveVanPersie: The IPA you added to
Brian Dabul is wrong. It's [ˈbɾaʝan], not [ˈbɾjan] because it's an English loanword - see
[6]. We've already discussed the issue of loanwords in Slovak, the same applies to Spanish.
The same applies to
Movistar, which is correctly transcribed [moβisˈtar]. /st/ is not a valid onset in Spanish (compare Spanish estar with Italian stare and Latin stāre), and the /s/ can be retracted to [h], which in most dialects is only possible in the syllable coda.
Mr KEBAB (
talk)
04:38, 19 January 2018 (UTC)
@
Awesomemeeos:[θẽə̃ŋks!hæpinʉjiɻtəjʉtʉ.ðəfaɪɻ̍wɻ̍kshiɻwɻ̍ɻiɫiɫæʊd‿ðɵʊ!wʌŋ‿kədθɪŋk"ɑɻðəɻʌʃn̩zətækɪŋʌs?"θẽə̃ŋk‿ɡɑd(oɻhʉɛvɻ̍/wʌɾɛvɻ̍jəbəɫivɪn)aɪhəvdisn̩thɛdfɵʊnz.ðɛɻtʃip,jɛʔvɛɻiəfɛktɪvəp‿bɫɑkɪŋæʊtʔæʊtsaɪdnoɪzəz,əspɛʃɫ̩iwɛnjəpɫeɪðɪs:[7].]Mr KEBAB (
talk)
01:40, 1 January 2018 (UTC)
Danish phonology
Hi, would you be able to review
this edit at
Danish phonology? Sources I quickly turned to all had /r/, /R/, or /ʁ/ as the phonemic representation of the rhotic. I'm not familiar with the topic but I highly doubt the edit is well founded, so I'd appreciate if you could take a look at it. Thanks.
Nardog (
talk)
23:03, 7 January 2018 (UTC)
@
Nardog: Done. Which source uses /ʁ/ by the way? The only phonemic symbol I've ever seen is /r/, but I haven't read a lot about Danish. It doesn't help that my Danish is poor (as is my Serbo-Croatian)...
Mr KEBAB (
talk)
02:04, 8 January 2018 (UTC)
The IPA illustration. I didn't have access to the full text; I just looked at the first page and saw /ʁ/ in the consonant inventory table. So it may be a different symbol used in the text.
Nardog (
talk)
04:32, 8 January 2018 (UTC)
@
Nardog: I'm pretty sure that the IPA illustration focuses on phones much more than phonemes, so it's probably [ʁ] (but I could be wrong). Basbøll certainly writes /r/. 04:53, 8 January 2018 (UTC)
@
Nardog: If Grønnum also lists [ð] and perhaps [ɪ̯,ʊ̯,ʌ̯] (which she changed to [j,w,ɐ̯] in newer works) in the consonant table then it's a table of phones, rather than phonemes.
Mr KEBAB (
talk)
14:02, 8 January 2018 (UTC)
Concerning MBMBaM
Aww see, you're right. I understood the concept of "allophones" but didn't know the terminology. I realize now what you meant. Sorry about the trouble. I won't undo your edits anymore. — Preceding
unsigned comment added by
Dabblequeen (
talk •
contribs)
16:32, 21 January 2018 (UTC)
This page is an
archive of past discussions for the period 25 December 2017 - 21 January 2018. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the
current talk page.
Alveolo-palatals
While I agree they don't deserve a column in the IPA consonants table, simply because the term is limited to fricatives and affricates so the column needlessly widened the already crowded table, they definitely have to be listed somewhere on one of the tables. Should they be moved back to the co-articulated consonants? Or just outside the pulmonics/affricates tables?
Nardog (
talk)
17:01, 25 December 2017 (UTC)
@
Nardog: It's not limited to fricatives and affricates. All consonants marked as palatal on the official IPA chart can be alveolo-palatal (apart from [cç,ɟʝ,ç,ʝ] and perhaps [j̊,j], but that may be debatable). [tɕ,dʑ,ɕ,ʑ] can stay where they are right now, because they're sibilant counterparts of [cç,ɟʝ,ç,ʝ]. The classification is correct.
Mr KEBAB (
talk)
17:19, 25 December 2017 (UTC)
Ah, you're right, my mistake. I also missed the fricatives and affricates were moved to the sibilant rows. I'm fine with the current layout. Sorry to have bothered you.
Nardog (
talk)
17:26, 25 December 2017 (UTC)
That would be consistent with other alveolo-palatals (those which do not have their own IPA symbols). But the thing with [(t)ɕ, (d)ʑ] is that they sound way closer to [(t)ʃ, d(ʒ)] than to [(c)ç, (ɟ)ʝ]. I wonder why only the fricatives are given their own IPA symbols, but that would be my best guess (and perhaps they're more common cross-linguistically?). And I would want to know why before we decide to merge the articles for [(t)ɕ, (d)ʑ] with [(c)ç, (ɟ)ʝ].
Nardog (
talk)
13:34, 26 December 2017 (UTC)
@
Nardog: That depends on what you consider [tʃ,dʒ,ʃ,ʒ] to be. IMO, considering canonical [tʃ,dʒ,ʃ,ʒ] to be weakly palatalized is just weird. It's like saying that canonical [l] is palatalized or that canonical [n] is velarized. The coloring should be neutral. Plus, no language contrasts weakly palatalized [tʃ,dʒ,ʃ,ʒ] with [tɕ,dʑ,ɕ,ʑ], but there are languages (Polish, Mandarin, etc.) that contrast hard, completely unpalatalized [tʃ,dʒ,ʃ,ʒ] with [tɕ,dʑ,ɕ,ʑ].
If you consider canonical [tʃ,dʒ,ʃ,ʒ] to be hard, then [tɕ,dʑ,ɕ,ʑ] are much closer to [cç,ɟʝ,ç,ʝ]. If you don't, they sound as something in between them, but a bit closer to [tʃ,dʒ,ʃ,ʒ] because of their sibilant nature.
Mr KEBAB (
talk)
14:08, 26 December 2017 (UTC)
@
Nardog: On second thought, considering [tʃ,dʒ,ʃ,ʒ] to be weakly palatalized is much weirder because these variants are much closer to [tɕ,dʑ,ɕ,ʑ] than the flat postalveolars. I hope the IPA clarifies this should they ever write the next edition of the Handbook.
Mr KEBAB (
talk)
15:15, 8 January 2018 (UTC)
@
Nardog: I'd put that on hold. We should merge the articles on [tʃ,dʒ,ʃ,ʒ] with the ones on [tʂ,dʐ,ʂ,ʐ] - see
User:Mr_KEBAB/VPAS (it's far from being 100% finished). It's a bit too early to start a discussion about that on
Wikipedia:WikiProject Linguistics, and I don't want time pressure to finish all of the four articles. If editors agree on the merge (there are good reasons for it), then we can think of appropriate article names, which will be a minor issue anyway.
Mr KEBAB (
talk)
18:50, 25 December 2017 (UTC)
As far as other postalveolar consonants are concerned, I'd also move them to the corresponding articles about retroflex consonants. Maybe the articles on [ɹ] and [ɻ] should be fully merged, I'm not sure.
Mr KEBAB (
talk)
18:52, 25 December 2017 (UTC)
I didn't realize you were prepping those articles, sorry if I stepped on your feet. But wouldn't you at least agree there need to be articles named "voiceless/voiced postalveolar fricative/affricate"? Your wording ("merge ... with the ones on [tʂ,dʐ,ʂ,ʐ]", "...move them to the corresponding articles about retroflex consonants") suggests you're intending to merge the postalveolars into articles named "...retroflex...", which seems odd. Don't you actually mean the other way around? Or do you actually think they should be gathered in articles that have "retroflex" in their names?
I also have a feeling it wouldn't be a bad idea to accumulate only truly retroflex (subapical) sounds into the articles on retroflexes, provided that they are common enough to warrant the articles, though I don't know how practical it is to ascertain the actual tongue shape of each sound described as retroflex.
Nardog (
talk)
13:34, 26 December 2017 (UTC)
No wonder you didn't, the only person I told about that was No such user. I only said that because the merge is very likely to occur, so we'll rename them again anyway.
I have no idea how we should call them. I'm counting on other editors to choose the names. Lazy, I know.
Ok, but the very reason I want to merge the articles on postalveolar/retroflex sibilants is that flat postalveolar sibilants found in Polish, Lithuanian etc. are sometimes called retroflex. The corresponding Mandarin sounds are always called retroflex, whereas the corresponding French sounds are never considered to be retroflex. But they're all the same sounds pronounced with flat tongue. (EDIT: That's irrelevant to what you wrote) As far as other retroflex consonants are concerned, that strikes me as severely limiting. What about non-subapical retroflexes/postalveolars that are transcribed with the retroflex symbols?
Mr KEBAB (
talk)
14:08, 26 December 2017 (UTC)
Frankly―and you seem to agree on this―the current organization of speech sounds is an utter mess. You've already touched upon this idea at
WT:LING but there are a lot of scarcely attested sounds that are better merged with related sounds. I think [
ä and [
ɒ̈ can merge too. There are also completely unattested
Labiodental trill etc., which I think should be redirects to an article called
Unattested sound or something to that effect (or to some existing article).
Nardog (
talk)
13:34, 26 December 2017 (UTC)
Yeah. I screwed up by not merging [ɪ̈,ʊ̈] with [ɘ,ɵ] a few months ago when I said I would. But the articles are almost ready and the merge can still be performed. I think that [ɯ̽] and [ɤ] can stay separate. At least I think that's what you're talking about.
I'd merge [ɒ̈] with [ɞ̞] but not necessarily [ä]. [ɶ,ɑ,ɒ] also cover both open and near-open sounds, but it's good to preserve the rounded-unrounded distinction, which is consistent with the rest of the articles.
Mr KEBAB (
talk)
14:08, 26 December 2017 (UTC)
Actually, now that I think about it, I'm not so sure if I would like "merging [ɪ̈,ʊ̈] with [ɘ,ɵ]". As far as English is concerned, [ɪ̈, ʊ̈] are better transcribed as [ɨ̞, ʉ̞] (or even simply [ɨ, ʉ]) because ⟨ɨ, ʉ⟩ aren't occupied by other phonemes. Are [ɪ̈, ʊ̈] known to contrast with [ɨ, ʉ] in any language? If not, I feel like they are better merged with [ɨ, ʉ] than with [ɘ, ɵ].
@
Nardog: The reason [ɪ̈,ʊ̈] should be merged with [ɘ,ɵ] is that [ɪ,ʊ] cover both near-close and close-mid vowels (see
near-close near-front unrounded vowel and
near-close near-back rounded vowel). Because of that, it's pretty much impossible to determine whether vowels transcribed with [ɪ̈,ʊ̈] (or the corresponding non-IPA symbols [ᵻ,ᵿ]) are near-close or close-mid unless that is explicitly stated in a given source. I agree that you should expect[ɪ̈,ʊ̈] to be near-close, but that's not good enough for Wikipedia.
There's also the issue of the centralizing diacritic itself, which some authors (e.g. Collins & Mees in their Phonetics of English and Dutch) use to mean mid-centralized, not centralized, or even both. For example, they say that the pre-/r/ allophone of /i/ is [ïə], but it's actually [ɪə] (at least in Randstad, other speakers probably use a fully close vowel with an optional schwa offglide). So even when we know that a scholar considers [ɪ,ʊ] to be near-close, we can't really be sure that their [ɪ̈,ʊ̈] are near-close as well. So it makes a lot sense to merge [ɪ̈,ʊ̈] with [ɘ,ɵ]. It's consistent with the series of edits I made to [ɪ,ʏ,ʊ] almost a year ago.
When we merge [ɪ̈,ʊ̈] with [ɘ,ɵ], we'll be also able to list more English dialects with central rounded FOOT and NURSE. This is good for our readers. Plus, the variants of FOOT transcribed with [ɵ] and [ʊ̈] may not be different at all (different sources use different notations, very often without providing vowel charts).
Hi, long time no see! I wonder what's your opinion on Polish (and maybe other Slavic) names pronunced in English? For example,
Łódź beling pronunced like
WOOCH? Do they sound strange or foreign? I'm very interested to hear your opinion. – they call meAWESOMEmeeos ...
[ˈɔɪ̯]!
12:09, 29 December 2017 (UTC)
@
Awesomemeeos:[wuːtʃ] for Łódź does sound foreign. It must be so, because only the first sound is correct as far as Standard Polish is concerned. But we're not talking about a non-native pronunciation of a Polish name in Polish, but an English approximation thereof. It's a big difference. I'd have no problem with hearing [ˈbɪdɡɑʃtʃ] or [ˈbɪdɡɑʃ] for Bydgoszcz and [ˌʃvɪnɵʊˈʉʃtʃə] for Świnoujście. Only someone that doesn't know how phonetics/phonology work would. But there's no excuse for not getting the stress right.
Mr KEBAB (
talk)
13:31, 29 December 2017 (UTC)
1) It seems that words in onomastika are proper names, but geographical names like the mountain Konjuh are also proper names... So I am not sure if they are given names and surname...
@
LoveVanPersie: I'm not sure if there's any English word that begins with /ər/ as opposed to /ɜːr/. It seems to me that only the latter is allowed in the unstressed word-initial position. @
Nardog:, can you confirm that? You have a better grasp of phonology than me. The non-rhotic pronunciation in the third link is definitely /ɜːrˈnɛstoʊˌɛskəˈbeɪdoʊ/, with strong /ɜːr/, not weak /ər/ (phonetically [əː] and [ə] in the narrator's accent, but you probably know that already). Bear in mind that in much of the North America there's no actual distinction between /ər/ and /ɜːr/. There's a related English name Ernest by the way, and it's pronounced /ˈɜːrnɪst/, with the same /ɜːr/ but stressed (/ər/ can't be stressed in accents that contrast it with /ɜːr/).
The correct respelling is ur-NES-toh ES-kə-BAYD-oh. Stressed /ɛ/ can never occur in the syllable-final position.
It's initial in both cases, but it seems to me that only the female pronouncer treats the first name as trisyllabic ([ˈsaːnɪaː]). The man seems to compress /-nɪ.aː/ into [-njaː], yielding [ˈsaːnjaː]. This is very similar to what happens to unstressed /i/ before /ə/ in English (see
Help:IPA/English).
1. For surnames, look under pr. I'm not sure what it stands for, but that's where they are. For first names, look under m. os. ime (for male first names) and ž. os. ime (for female first names).
2. They're two different surnames. Hers is Martić with ć.
3. Jȕrak is diminutive of Juraj. The surname is Jȕrāk, with a long post-accentual a.
4-6. Both pronunciations are correct.
7. I'd only transcribe Bȁšić. I don't know about the other variant. It could be that Elly Bašić insisted on pronouncing her surname Bàšić like
Rachel Weisz insists on the /vaɪs/ pronunciation of her surname. Maybe it has dialectal connotations. These are all speculations by the way. Mail HJP if you want a definitive answer.
You have a better grasp of phonology than me. Well, that's flattering but probably not true. Anyways, I think you're right, I can't think of any instance of word-initial /ər/. Searching in CEPD/LPD seems to confirm that. Happy new year.
Nardog (
talk)
06:05, 4 January 2018 (UTC)
LoveVanPersie That's almost correct. The IPA is /θəˈnɑːsiˌkɒkɪˈnɑːkɪs/ (thə-NAH-see KOK-ih-NAH-kis). I've noticed that you're still confused about the lax vowels, which can't ever end stressed syllables (save for extremely rare instances of loanwords such as pho). You don't have to use the IPAc-en and respell templates here by the way. You're making an effort that doesn't make a difference (I don't mean to sound disrespectful, I'm just trying to save your time). The reason we use the IPAc-en template in articles is because of the
mouseover feature and because it links to
Help:IPA/English where all of the symbols are explained. In the case of the respell template, we use it mainly because it links to
Help:Pronunciation respelling key.
Yes, the spelling itself proves it. Catalan doesn't have ⟨ñ⟩, they use ⟨ny⟩ instead (interestingly, that feature is shared with some African languages). Compare Catalan Catalunya[kətəˈɫuɲə] with Spanish Cataluña[kataˈluɲa].
Mr KEBAB (
talk)
23:53, 4 January 2018 (UTC)
The correct syllabification seems to be /ˈkVr.i.ɒs/ (where V stands for vowel), so /ˈkiːriɒs/ is probably impossible. The only possibilities are /ˈkɪriɒs/ and /ˈkɪəriɒs/ (Australian [ˈkiɹ-] and [ˈkiːɹ-], respectively), and I think it's the former. The dialect of English I speak ("I'm trying to imitate" is probably a more suitable phrase) doesn't have phonemic vowel length, so it's a bit hard for me to distinguish Australian /ɪ/ from Australian /ɪə/, as they have pretty much exactly the same quality and differ only in length. Strangely, I don't have such problems with English/ɪ/ and /ɪə/ in the same positions. Apparently I still have more to learn as far as AuE is concerned. /ɪ/-tensing still sounds strange to my ears.
Here, the pronunciation with /ɪr/ is more obvious, as the commentator is British. I'm 95% sure that the correct IPA is /ˈkɪriɒs/, with the lax /ɪ/.
Mr KEBAB (
talk)
01:01, 5 January 2018 (UTC)
@
LoveVanPersie: Nope, unstressed syllable-final /ɒ/ appears only before consonants. The correct IPA is /ˈtrɔɪsi/ (TROY-see). The [ɪ] is definitely non-syllabic ([ɪ̯]) and short.
Mr KEBAB (
talk)
01:26, 5 January 2018 (UTC)
Sorry to nitpick, but /iː.r/ is certainly possible albeit rare, most notably in some Americans' pronunciation of hero and zero (see AoE pp. 481–2). Parallels to this are also found in Nero, pharaoh, and guru.
Nardog (
talk)
02:56, 5 January 2018 (UTC)
Ah, that's what you meant. I agree the name does seem to have MIRROR, at least judging from the UK pronunciation in the video you linked to.
Nardog (
talk)
03:25, 5 January 2018 (UTC)
@
LoveVanPersie: 1. Yes, but only when there's a pause between the words, as on the recording. The correct IPA for the variant without a pause is [ˈnorberdˈɡomboʃ] (notice that ⟨ɡ⟩ is the correct transcription, not ⟨g⟩).
2. Yes.
3. Yes.
4. No, it's /ˈstɛfɑːnˈkɒzlɒv/. What we transcribe /ɑː/ on
Help:IPA/English can never be written with ⟨o⟩.
5. Visit
Help:IPA/Slovak and
Slovak orthography and tell me the IPA. It's straightforward (remember that the surname is written Kovalík, with ⟨í⟩).
6. The guy has a non-native accent and so the recording can't be used as a source.
@
LoveVanPersie: Our
diaphonemic transcription forces us to write /ˈkɒzlɒv/. From it, you can easily deduce that the General American pronunciation is [ˈkɑːzlɑːv].
"Consonant clusters containing both voiced and voiceless elements are entirely voiced if the last consonant is a voiced one, or voiceless if the last consonant is voiceless." But pr and kr seem exceptions?
LoveVanPersie (
talk)
06:02, 11 January 2018 (UTC)
@
LoveVanPersie: I'm not knowledgeable enough to discuss Slovak phonology with you. Try to get your hands on The Lexical Phonology of Slovak which should answer your question. Perhaps
sonority hierarchy will answer that question too.
And should we need to remove the tie bars on
Help:IPA/Mandarin, where [t͡ɕ], [t͡ɕʰ], [ʈ͡ʂ] and [ʈ͡ʂʰ] are with tie bars but [ts] and [tsʰ] are without them. In
Template:IPAc-cmn, transcriptions for the six affricates are all left out tie bars.
Is the /v/ in Viktória pronounced [v] or [ʋ]? I think it's a normal situation so it is pronounced [v]. But it seems a case of "[ʋ] occurs in all other cases". So I'm confused...
Is the /m/ in Škamlová pronounced [m] or [n]? I think it's the former according to
Slovak phonology, in which /m/ only "has the allophone [ɱ] in front of the labiodental fricatives /f/ and /v/".
Is Rebecca Šramková pronounced [ˈrebet͡sa ˈʃraŋkoʋaː]? And is
Šramková a Czech or Slovak surname?
@
LoveVanPersie: The answer is on
Help:IPA/Slovak, which you seem to be quoting. The note you're quoting is unambiguous, so I don't understand why you're saying that it's a normal situation so it is pronounced [v] (whatever you mean by normal situation). There's nothing to be confused about.
/m/never assimilates to [n] or [ŋ] in standard Slovak. It'd be a strongly abnormal pronunciation that doesn't occur in any language I'm aware of.
Please don't ask me about the pronunciation of Slovak words.
Radoslav Ivan is a better candidate for that, but bear in mind that I am somewhat abnormally generous when it comes to answering your questions. Radoslav might not have that much time or patience.
Once again, you're asking me a question you could've answered yourself. See
Šramková and pay attention to the spelling.
Also, are you 100% positive that the first name of
Chantal Škamlová is pronounced with [x], not [ʃ]? It seems to be a French name. The same applies to
Andrej Martin - are you sure that the surname is pronounced with [t] and not [c]? The research we did was on a first name, not a surname. They don't have to be pronounced in the same manner (see
homograph). What I'm saying is that you should never add IPA if you're not sure if it's correct. It could be taken as
WP:VANDALISM and get you banned.
Mr KEBAB (
talk)
19:30, 12 January 2018 (UTC)
@
LoveVanPersie: But you should've expected that. Hardly any language respells every single loanword. The Serbian variety of Serbo-Croatian does, Slovak doesn't. It doesn't even have a fully phonemic orthography.
Mr KEBAB (
talk)
15:58, 14 January 2018 (UTC)
@
LoveVanPersie: The first name is pronounced /bərˈnɑːrdə/, but I'm not sure about the surname. GA doesn't distinguish /ɛr/ from /ɛər/, so it's either /ˈpɛrə/ or /ˈpɛərə/. It'd be perfect if I could hear the British pronunciation of the surname.
Mr KEBAB (
talk)
09:34, 18 January 2018 (UTC)
@
LoveVanPersie: The IPA you added to
Brian Dabul is wrong. It's [ˈbɾaʝan], not [ˈbɾjan] because it's an English loanword - see
[6]. We've already discussed the issue of loanwords in Slovak, the same applies to Spanish.
The same applies to
Movistar, which is correctly transcribed [moβisˈtar]. /st/ is not a valid onset in Spanish (compare Spanish estar with Italian stare and Latin stāre), and the /s/ can be retracted to [h], which in most dialects is only possible in the syllable coda.
Mr KEBAB (
talk)
04:38, 19 January 2018 (UTC)
@
Awesomemeeos:[θẽə̃ŋks!hæpinʉjiɻtəjʉtʉ.ðəfaɪɻ̍wɻ̍kshiɻwɻ̍ɻiɫiɫæʊd‿ðɵʊ!wʌŋ‿kədθɪŋk"ɑɻðəɻʌʃn̩zətækɪŋʌs?"θẽə̃ŋk‿ɡɑd(oɻhʉɛvɻ̍/wʌɾɛvɻ̍jəbəɫivɪn)aɪhəvdisn̩thɛdfɵʊnz.ðɛɻtʃip,jɛʔvɛɻiəfɛktɪvəp‿bɫɑkɪŋæʊtʔæʊtsaɪdnoɪzəz,əspɛʃɫ̩iwɛnjəpɫeɪðɪs:[7].]Mr KEBAB (
talk)
01:40, 1 January 2018 (UTC)
Danish phonology
Hi, would you be able to review
this edit at
Danish phonology? Sources I quickly turned to all had /r/, /R/, or /ʁ/ as the phonemic representation of the rhotic. I'm not familiar with the topic but I highly doubt the edit is well founded, so I'd appreciate if you could take a look at it. Thanks.
Nardog (
talk)
23:03, 7 January 2018 (UTC)
@
Nardog: Done. Which source uses /ʁ/ by the way? The only phonemic symbol I've ever seen is /r/, but I haven't read a lot about Danish. It doesn't help that my Danish is poor (as is my Serbo-Croatian)...
Mr KEBAB (
talk)
02:04, 8 January 2018 (UTC)
The IPA illustration. I didn't have access to the full text; I just looked at the first page and saw /ʁ/ in the consonant inventory table. So it may be a different symbol used in the text.
Nardog (
talk)
04:32, 8 January 2018 (UTC)
@
Nardog: I'm pretty sure that the IPA illustration focuses on phones much more than phonemes, so it's probably [ʁ] (but I could be wrong). Basbøll certainly writes /r/. 04:53, 8 January 2018 (UTC)
@
Nardog: If Grønnum also lists [ð] and perhaps [ɪ̯,ʊ̯,ʌ̯] (which she changed to [j,w,ɐ̯] in newer works) in the consonant table then it's a table of phones, rather than phonemes.
Mr KEBAB (
talk)
14:02, 8 January 2018 (UTC)
Concerning MBMBaM
Aww see, you're right. I understood the concept of "allophones" but didn't know the terminology. I realize now what you meant. Sorry about the trouble. I won't undo your edits anymore. — Preceding
unsigned comment added by
Dabblequeen (
talk •
contribs)
16:32, 21 January 2018 (UTC)