Talk:No audible release

Only stops?
I've just edited the Korean language article:
 * Only seven consonant allophones are found at the end of syllables: and . They are all unreleased.

and are certainly stops, whether they be oral or nasal stops. But I don't think I've ever heard /l/ being mentioned as a stop. But fianl /l/ sounds unreleased to me, at least in Korean and English. It sounds released to me in the French pronunciation of the word elle.

So I want to ask if unreleased laterals are also possible. Is that true?

Am I onto something, or am I confused? --KJ 06:21, 9 June 2006 (UTC)


 * The phonemes are not normally considered stops. Release is only meaningful for a stop. If there is no occlusion, how can it be released? In French elle, there is a shwa after the l.  &minus;Woodstone 13:44, 9 June 2006 (UTC)


 * Thank you. I was confused about the occlusion bit, and was just thinking about "articulators staying in touch" as in Korean. --Kjoonlee 14:20, 27 June 2006 (UTC)


 * There is no schwa after the l in elle, http://en.wiktionary.org/wiki/elle#French Perhaps you are thinking of Italian ella. 99.236.215.170 (talk) 18:40, 13 September 2014 (UTC)
 * There is, it's just non-phonemic. That's what (s)he means by the term "released". Kbb2 (ex. Mr KEBAB) (talk) 15:55, 4 April 2020 (UTC)

Placement of the diacritic
The diacritic is supposed to follow the consonant, not be placed on top of it. See the IPA chart here: http://www.langsci.ucl.ac.uk/ipa/diacritics.html — Preceding unsigned comment added by 142.151.171.95 (talk) 13:33, 24 May 2013 (UTC)
 * No, it's a combining character. It comes (or should come) after a d cos there's nowhere else for it to go, I suppose. It'd look ridiculous on top of the ascender. — Lfdder (talk) 13:52, 24 May 2013 (UTC)
 * Surely the important thing is the way phoneticians use it, not whether Unicode classifies it as a combining character? If you do a Google Books search for "unreleased stop" you'll see many, many examples in which the diacritic follows the consonant (including consonants with no ascender, such as [p] and [g]). Here are some examples etc etc. In fact, the prominent British phonetician John Wells notes in this blog post that "Unfortunately, U+031A doesn’t display properly". He substitutes another character that has the same shape but appears following the consonant. And finally, Pullum and Ladusaw's Phonetic Symbol Guide says that the diacritic occurs "following a stop consonant symbol" (p. 239). I think all of the above is sufficient to illustrate that the wiki article is incorrect! 142.151.171.95 (talk) 22:05, 25 May 2013 (UTC)
 * (The comments from 142.151.171.95 are from me; I'd forgotten about my long-disused account.) WillNL (talk) 22:10, 25 May 2013 (UTC)
 * One last note: even the Unicode standard shows the diacritic following the symbol (search for 031A in this pdf). It seems that the standard has been incorrectly implemented in certain fonts, including the one that wikipedia uses, such that the diacritic appears directly above the symbol rather than above and to its right. WillNL (talk) 22:14, 25 May 2013 (UTC)

The problem isn't Unicode, but the font you're using. Evidently Wells had a badly designed font as well. Gentium displays just fine. — kwami (talk) 16:24, 28 May 2013 (UTC)

Palatal nasal before consonants
I think that preconsonantal palatal nasals are typically unreleased. For instance, in Polish, the standard pronunciation of wziąć ('to take') is. I was quite surprised to learn that (some) native speakers don't hear it as a variant of (then again, most of them actually think that $⟨ą ę⟩$ are pronounced as nasal vowels in all positions, which is obviously false. Hence the extremely common mispronunciation, which does contain a nasal vowel because of the epenthetic fricative). Kbb2 (ex. Mr KEBAB) (talk) 13:53, 3 April 2020 (UTC)

The same goes for the palato-alveolar nasal in English crunch and in Spanish poncho  'poncho'. If they were released into a vowel, we'd hear a weak after the nasal. Kbb2 (ex. Mr KEBAB) (talk) 09:24, 6 April 2020 (UTC)
 * Maybe I'm misunderstanding something, but I was under the impression that a release was a key feature of plosives like, not sonorants like . There might be a more sophisticated investigation of this through sourcing, but I would imagine that consonant clusters in general probably feature more unreleased plosives. — Æµ§œš¹  [lɛts b̥iː pʰəˈlaɪˀt] 16:26, 6 April 2020 (UTC)
 * Try whispering and you'll see that they're released before a vowel. There's a very weak plosion that resembles . It's most obvious in the case of the velar nasal.
 * The way I see it, the key feature of the release portion of is the palatal approximant . If it wasn't there, you'd get what sounds like a retroflex nasal . Dutch oranje , Polish spanie   and Spanish año   (EDIT: IPA fixed - see below) all feature and can't not feature a palatal approximant between the nasal and the following vowel, and by that it massively differs from the corresponding voiced fricative  (compare the alveolar pair , which has no effect on the following vowel). Whether it's as strong and long as the  phoneme in those languages is up to debate for sure, but it's there. We should find a source that says that and put that information in palatal nasal.
 * Yes, I know that fricatives can't be unreleased. What I mean is that can be "released" (not really) directly into a vowel, without an intervening . Kbb2 (ex. Mr KEBAB) (talk) 05:38, 7 April 2020 (UTC)
 * If what I'm saying is true then it perfectly explains why native speakers of English and German approximate foreign as . It also explains the spelling $⟨ny⟩$ in Catalan, Hungarian and many African languages. Kbb2 (ex. Mr KEBAB) (talk) 05:50, 7 April 2020 (UTC)
 * I was going to say the same thing as Aeusoes1. And I think he is saying that the notion of the dichotomy between released and unreleased (or inaudibly released) usually only applies to plosives, not that nasals are never released.
 * Occlusives are typically broken into three phases: approach, closure (aka occlusion/hold), and release (aka plosion/burst), and in plosives and nasals, only the closure is requisite. Between every pair of segments (or a segment and prosodic boundary) there is a transient stage where articulators move and various configurations may change (called onglides and offglides, not to be confused with ones in diphthongs), but whether that stage should be treated as a separate segment is almost always a phonological, not phonetic, question. Nardog (talk) 06:00, 7 April 2020 (UTC)
 * Then palatal nasals seem to be an exception to that. In Polish (at least in the prescriptive standard), we even have minimal pairs like dania 'dishes' vs. Dania  'Denmark'. But the first word clearly has a  too (proven by how different it sounds in comparison with  in pięć  'five', which sounds just like the hard ), it's just that it's shorter and non-phonemic. Whether you transcribe the word meaning 'dishes' as  or  matters very little, you can even write it . The contrast between  and  manifested as the absence vs. presence of  (corresponding phonemically to, not to any non-phonemic glides) manifests phonetically as such only before , as in the near-minimal pair Ani  'Annie' (gen. sg.) vs. Danii  'Denmark' (gen. sg.). Elsewhere, the 'palatal transition' between the nasal and the vowel is essential for a native speaker to recognize the nasal as  and not  (without it, dania 'dishes' would be understood as dana 'given'). Notice how hard it is to directly release  into a vowel anyway. You almost have to force it, and if you succeed at it the result sounds postalveolar (between dental/alveolar and retroflex) and so not soft at all. I'm beginning to think that  by definition has a diphthongizing effect on the following vowel (also across word boundaries).
 * Should it be transcribed? Yes, it should - at least in Polish. In words such as słońce 'sun' (which, I guess, would normally be phonemicized as ) the nasal is phonetically dental (!) and its softness is expressed only by diphthongizing the preceding vowel: . If this diphthongization is transcribed, why should dania (lowercase) and Ania 'Annie' be transcribed as anything other than and  (or the equivalent transcriptions proposed above)? Unless it's not important to maintain consistency in the level of narrowness of phonetic transcription - then, I'd say, anything goes.
 * In Slovak, the sequences of a palatal nasal + vowel in words such as Daniela (female name) are phonemicized as + rising diphthong: . Here, the speaker clearly says, with a non-standard (see Kraľ 1988 "Pravidlá slovenskej výslovnosti") sequence of  plus a mid vowel (though mid central rather than mid front) instead of a diphthong . If turning  into a sequence of  +  is an actual pronunciation error (and it is according to Kraľ) then I don't see how  can differ from , especially if  is disallowed after  (, can you help me here? Can the plain, undiphthongized  follow the palatal nasal in Slovak?). The distinction between  (assuming that it really isn't , per Kraľ),  and  seems to me hard or even almost impossible to maintain as the contrast between the latter two sequences would sooner or later collapse to , unless  developed into the retroflex . Kbb2 (ex. Mr KEBAB) (talk) 11:35, 8 April 2020 (UTC)
 * It seems to me that the main point is that when you're producing (at least the alveolo-palatal variant), the positioning of the tongue is such that it's very difficult not to produce the offglide. Remember that retroflexes are apical corono-(pre)palatal (rather than dorso-palatal like cardinal ), so the difference between  and the alveolo-palatal  is apical vs. laminal. Perhaps  could be seen as a "nasal affricate" of sorts that gets deaffricated before, with its nasal stop equivalent being ? Kbb2 (ex. Mr KEBAB) (talk) 06:44, 12 April 2020 (UTC)

"Hundred pounds"
I can't make sense of the following (line 11): "For example, hundred pounds may sound like [ˈhʌnd͡ʒɨd̥̚ ˈpʰaʊnd͡z] but X-ray and electropalatographic studies demonstrate that since inaudible and possibly-weakened contacts may still be made, the second /d/ in hundred pounds does not entirely assimilate a labial place of articulation but co-occurs with it." If the sentence was saying that listeners perceive [ˈhʌnd͡ʒɨb̚] ([b] substituted for [d] as a result of assimilation) then the rest of the sentence would make sense to me. Is it possible that the example is wrong? RoachPeter (talk) 09:41, 10 November 2021 (UTC)
 * You're right, it was but it was "corrected" earlier this year. Thanks for spotting this. Nardog (talk) 10:36, 10 November 2021 (UTC)