Talk:International Phonetic Alphabet/Archive 14

Archive 10

←

Archive 12

Archive 13

Archive 14

Unicode character U+1AC8

Now that it's been encoded, should we mention the Unicode character U+1AC8 COMBINING PLUS SIGN ABOVE ⟨◌᫈⟩ in the Diacritics and prosodic notation section as well as in article Phonetic symbols in Unicode#Diacritics? Love —LiliCharlie (talk) 04:36, 17 October 2021 (UTC)

I don't think so. That was encoded for non-IPA use. When an IPA plus or minus doesn't fit under a letter, the solution is to place it after the letter as a spacing diacritic, the same as we do with the tacks. The minus, of course, couldn't be placed above, as that would make it a macron, and it would make little sense to do that for the plus but not the minus. — kwami (talk) 04:49, 17 October 2021 (UTC)

That's nonstandard since 1989 though (Handbook, pp. 173, 183). I don't understand why "it would make little sense to do that for the plus but not the minus". After all we list ⟨◌̊⟩ for voiceless but not ⟨◌̌⟩ for voiced because it interferes with rising tone. Unicode also lacks superior tacks (except for up at U+1DF5, which was added for the pre-IPA Webster notation in 2017), and in a strict reading of the IPA chart and Handbook, Unicode doesn't provide a way to avoid interference with a descender for ⟨◌̠ ◌̞ ◌̘ ◌̙⟩. Nardog (talk) 05:53, 17 October 2021 (UTC)

Before Kiel, the tacks and plus/minus could come after the letter even when there was no descender. Part of the goal of Kiel was resolving duplicate symbols like that. I don't know why the Handbook would still give the option for raised/lowered: they're not even on letters with descenders. Unless the parentheticals are merely explanatory? I'd need to dig into the resolutions. — kwami (talk) 07:46, 17 October 2021 (UTC)

I don't think identical diacritics for voiced and rising tone are ever ambiguous in real-world transcriptions, because that seems to imply that either voiceless obstruents can be the nucleus of a tone-carrying syllable, or that voiced sounds can be further voiced, and that would be astonishing. Such usage probably doesn't even violate the IPA's Principles, which are about symbols for sounds rather than symbols for sound features, and which also say (principle 7) that "[a] transcription always consists of a set of symbols and a set of conventions for their interpretation." (Italics are mine.) Love —LiliCharlie (talk) 21:46, 17 October 2021 (UTC)

I can certain imagine that someone might use it. But do we have any indication that anyone ever has? E.g. in typeset docs that wouldn't rely on Unicode. — kwami (talk) 18:30, 18 October 2021 (UTC)

I have no evidence for that, but the world is huge, and so is the number of languages specialist linguistic literature is published in. — It would be nice to have an HTML version of the Handbook (in full or in part) published on Wikisource, but the current licence doesn't allow that, I'm afraid. Love —LiliCharlie (talk) 20:09, 18 October 2021 (UTC)

I'd certainly have no problem understanding it if someone were to write [y] with a plus on top, so maybe we should mention it. But I don't know how we'd answer if someone asked about the minus. — kwami (talk) 03:22, 27 October 2021 (UTC)

Kirk Miller's Unicode request for phonetic punctuation & diacritics of January 2021 offers examples of plus sign above IPA letters with a descender in narrow phonetic transcriptions, for instance ⟨ɡ᫈⟩ and ⟨ŋ᫈⟩ on p. 6, fig. 8. — There is no COMBINING MINUS SIGN ABOVE in Unicode. If there was one, its glyph would probably be shorter and thinner than the macron in fonts like Charis SIL, as for the corresponding subscript diacritics ⟨◌̠⟩ vs. ⟨◌̱⟩. Love —LiliCharlie (talk) 08:04, 27 October 2021 (UTC)

Sidenote: Before the Kiel Convention, when modifier/spacing minus was still in the IPA chart, it used to have serifs, i.e., it had a shape resembling character U+1D129 ⟨𝄩⟩ in the Unicode chart Musical Symbols. This glyph may have been chosen to distinguish the minus sign from the hyphen that was extensively used to denote a syllable break (now symbolized by ⟨.⟩), see for instance chapter XXXII "Syllable Separation" in Daniel Jones's An Outline of English Phonetics. Nowadays the hyphen is still in frequent use to replace repeated transcription signs, as when John Wells transcribes became as [bi ˈkeɪm bə‑] in the LPD, where he certainly doesn't mean a retracted schwa (and he explains that in the Other symbols list just before The Dictionary A–Z). Love —LiliCharlie (talk) 11:17, 27 October 2021 (UTC)

Unicode has separate code points for the macron below and the minus sign below, so naturally the answer is that it's not supported by Unicode. It is absolutely wild to me that:

• Despite the JIPA editor (Ian Maddieson) noting that "Since some pairs of diacritics have the same shape but differ only in their placement above or below the letter, acceptance of this proposal creates a few potential problems that may need to be addressed in a future vote" when they approved the "possibility of placing a diacritic above a symbol when that symbol includes a descender" in 1993, all they have done to address them to this day is added the word "Some" at the beginning of the note "Diacritics may be placed above..." in the 2015 chart.
• No placement variant except the ring above is assigned a dedicated IPA Number—not even the tie bar below, which does appear on the chart (it is inexplicably given, in parentheses, the same number as the non-combining undertie, on pp. 163, 203).
• Unicode doesn't encode superior variants of ⟨◌̠ ◌̞ ◌̘ ◌̙⟩ despite the note on the IPA chart.

Nardog (talk) 12:19, 28 October 2021 (UTC)

Okay, let's add it then.

If someone wanted the minus above, maybe they could add serifs like the old spacing mark. As long as it wasn't used in the same transcription as a macron, that should be enough to distinguish them.

If anyone has examples in print of ⟨◌̠ ◌̞ ◌̘ ◌̙⟩ above a letter, please share. We can certainly get them into Unicode for ppl to use. The spacing minus is clearly a problem with the hyphen, so I say we remove it from this article just as the IPA has removed it. But the tacks all have spacing variants now, and those are not ambiguous. The spacing ATR/RTR tacks were used as equivalent to plus/minus, but they'd work as ATR/RTR for standard IPA. — kwami (talk) 21:44, 28 October 2021 (UTC)

@Kwamikagami: Lass (1990) uses the advanced diacritic above ⟨ɑː⟩. Furthermore, Pavlik (2004) (from p. 87 onwards) has examples of the lowering diacritic used above the IPA letter, as well as one instance of the raising diacritic used in the same manner. Sol505000 (talk) 00:19, 29 October 2021 (UTC)

Nice find! I don't have access to the Lass -- is it just the + that's already in Unicode, or did you mean the ATR diacritic? But Pavlik -- they propose quite a few. Quite handy. I'll see what I can do. — kwami (talk) 03:32, 29 October 2021 (UTC)

@Kwamikagami: Sorry, I misread your request. It's the plus sign, not the [+ATR] diacritic. See for yourself, if you can access it: [1]. Sol505000 (talk) 21:08, 29 October 2021 (UTC)

Thanks. Odd they'd use the over-plus without a descender. But it's your 2nd source that's interesting re. additional support: quite a few IPA diacritics placed over letters with descenders. — kwami (talk) 01:57, 30 October 2021 (UTC)

Lass probably meant to put an additional (raising) diacritic underneath the letter. He describes Cape Town English /ɑː/ to be "characteristically less than fully open", like the LOT vowel. The two are described in a manner that strongly suggests they can be regarded as a short-long pair, as in Geordie (tho Wells himself considers them to be a short-long pair in RP). Sol505000 (talk) 03:56, 30 October 2021 (UTC)

@Kwamikagami: Lass (1984:158) uses the plus and the minus sign above ⟨r⟩ in and below chart 7.68, and he explains that means "advanced alveolar trill" vs. "a retracted one". Love —LiliCharlie (talk) 14:35, 30 October 2021 (UTC)

Luciano Canepari (1983): Phonetic Notation / La notazione fonetica has a ⟨k⟩ with a subscript macron and a superscript left-pointing tack three times in that book, on pp. 164 (English text), 169 (Italian text), and 173 (figure 8.1–2). If by spacing, subscript, as well as superscript ⟨꭪⟩ Canepari doesn't mean ATR but advanced (as in older IPA versions), it's still a sign that's used and should get a Unicode codepoint. And on p. 260 of that book there's another one: an ⟨ᴉ⟩ with a superscript down-poining tack for lowered. Love —LiliCharlie (talk) 15:36, 30 October 2021 (UTC)

@Sol505000 and LiliCharlie: Thanks! That's all very useful. And there are other things in Lass, such as ɰ with an overstruck tilde, which would require separate Unicode support. Probably want to stay away from canIPA, though.

While we're at it, have either of you seen ligatures for affricates like tθ dð tɬ dɮ (or for any non-coronal affricates), or turned β ð for approximants? I've seen proposals for the latter, but no actual use, and one source (Elderkin) for tɬ dɮ ligatures. — kwami (talk) 20:21, 30 October 2021 (UTC)

Canepari (1983) isn't ᶜᵃⁿIPA yet, though the vocoid system already is, apart from a few symbols. ᶜᵃⁿIPA proper came later and doesn't require a diacritic for advanced [k]; it has separate symbols for palatal, postpalatal, prevelar, "provelar" and velar contoids. These correspond to front, front-central, central, back-central, and back vocoids. Love —LiliCharlie (talk) 23:46, 30 October 2021 (UTC)

@LiliCharlie: Unfortunately, I don't have a copy of Canepari 1983. Could you quote the context of one or more of the instances of the k with the tack? I might be able to see a snippet in GBooks, which would be good enough. — kwami (talk) 07:58, 31 October 2021 (UTC)

In case GBooks can cope with bilingual books with numerous interspersed unusual letters and marks:

P. 164: "Therefore, they can freely arrange for the (labial) shape of the following phone, for instance round or spread: ". This is followed by a square bracket-enclosed transcription of 15 CV syllables, the last one is [ki] with the ⟨k⟩ modified by a subscript macron that symbolizes spread lips as well as the superscript left-pointing tack for advanced stricture location.

P. 169: "perciò possono liberamente predisporsi per la forma (labiale) del vocoide che li segua, per es. tonda o stesa: ". Followed by the same.

P. 173: "F 8.1-2 Coarticulatory modifications/Modificazioni coarticolatorie".

P 260 contains nothing but Canepari's transcription of the 140 recordings on the audio cassettes, so that is quite a challenge for the algorithm:

• "R 15 (b_ d_ ɡ_) i [ˈbi ˈdi ˈɡi] &c ɪ e ᴇ ɛ æ
• R 16 (b_ d_ ɡ_)", all with opentail ⟨ɡ⟩, followed by six more or less unusual symbols for front-central vocoids, the first one is the ⟨ᴉ⟩ with superscript down-pointing tack. Love —LiliCharlie (talk) 10:21, 31 October 2021 (UTC)

Thanks! The Italian worked, so we have evidence for three of the four tacks. — kwami (talk) 18:54, 31 October 2021 (UTC)

I believe it is unlikely to come accross a superscript right-pointing tack for RTR, as ⟨y⟩ is the only vowel sign with a descender and RTR consonants are usually written with the pharyngealization modifier ⟨ˤ⟩. Perhaps we can find one in very old literature where it modifies primary stricture location. Love —LiliCharlie (talk) 16:01, 1 November 2021 (UTC)

β̞

Section Diacritics lists β̞ as an example, with a link to Raising (sound change). Now that page defines its meaning by the tongue position. Is that really the case for β̞? When pronounced as a fricative, the tongue position makes no significant difference IMHO. I therefore take β̞ to indicate a lowered chin (or lower lip) position, that is, a more lax pronunciation, rather like an approximant. (Or even the approximant itself, given that it has no letter in IPA.) If that is so, then we need to adjust the article Raising (sound change). Otherwise, what's the point of using this particular consonant as an example – is there any example for when that distinction is significant? ◅ Sebastian 12:49, 15 November 2021 (UTC)

Sorry, that was a screwy link. Perhaps it was once linked to an article on 'lowered', but that article was merged into 'raised' and then had most of the 'lowered' stuff edited out as off-topic. I removed the link for now. Maybe someone can find a better one, or create a stub to explain it.

[β̞] is a bilabial approximant. It's notable because it occurs in Spanish. [ð̞] is a dental approximant occurs in Spanish, Danish and a number of other languages. — kwami (talk) 21:41, 15 November 2021 (UTC)

Thanks, kwami, for the explanation and the helpful examples, which are indeed notable because they occur in at least one of the word's most spoken languages.

But with all due respect, IMHO the problem is not the juxtaposition of high and low. Covering both opposites in one article makes sense. Nor was there a problem with the link. Removing it merely sweeps the problem under the rug. The problem seems to be that the down tack is used for other purposes than “lowering” as defined in the article on the topic. If that article is correct, then we need to change the description of the down tack here. Or else, “raising/lowering” can mean more than just the position of the tongue, in which case we need to change that article. Wouldn't you agree? ◅ Sebastian 09:24, 22 November 2021 (UTC)

@SebastianHelm: I'm not aware of any other uses. What were you thinking of? Lowering the tongue at the place of articulation has the effect of going downward in the IPA charts, e.g. vowels i > e > ɛ or consonants ɡ > ɣ > ɰ. You could even bring them together in ɟ > ʝ > j > i > e > ɛ. (Indeed, the IPA used to put the vowels in the same chart as the consonants, so that those series were more obvious.) — kwami (talk) 11:45, 22 November 2021 (UTC)

Thanks, that's interesting. While each of the connections you indicate with ‘>’, including ‘j > i’, makes sense, it never occurred to me to connect the dots and regard them as one “stricture series”, as the other article you bring up below calls it. ◅ Sebastian 15:33, 22 November 2021 (UTC)

I added a link to Relative articulation#Raised and lowered years ago, but someone deleted it because they apparently didn't understand the connection. So that's back, and I'll use it as the link in this article as well. — kwami (talk) 11:54, 22 November 2021 (UTC)

That solves the riddle! Your addition of “distinguish” is just what I meant with “other purposes than ‘lowering’ as defined in [Raising (sound change)]”. I just hadn't been aware that we already have an article about that, but I now see that, in section Raised and lowered consonants, it even explicitly mentions [β̞ ]. I see that you also added the appropriate link, so my original question is solved now. Thanks!

There only remain some minor issues, but they're not about this article: Such as whether other articulators are included in the second article, too – it currently only speaks of “tongue or lip”. (I know too little to word my question well; certainly the distinction between place and articulator ceases to make sense at some point; at the very least for glottal sounds.) Also, I don't understand yet why linguists need to reuse the terms “raise” and “lower” when they already have more than enough terms expressing basically the same distinction: “stricture”, “degree of contact”, “lenition” and “fortition”. ◅ Sebastian 15:33, 22 November 2021 (UTC)

'Raise' and 'lower' is convenient for vowels, where there is no real stricture, and the same diacritics are used for consonants, so I suppose it's natural to use the same terms. Also, lenition and fortition can involve other things. An example in the stricture table is ɬ > l, which involves voicing in addition to lowering. A lenited [t] might be [θ] or [s], or [d] or [ɾ], etc. And fortition of [h] might get you [p] (e.g. in Japanese). So e.g. you might need describe lenition in a particular case as lowering + voicing or sibilation or flapping.

Anyway, thanks for catching the problem. — kwami (talk) 21:46, 22 November 2021 (UTC)

Thank you, always a pleasure chatting with you. I now can see a motivation for multiplying terms. I could go on, but will restrain myself, since we've left the territory of this article now. As for the question about “tongue or lip”, I think I'll carry that over to Talk:Relative articulation some day. ◅ Sebastian 22:57, 22 November 2021 (UTC)

Principles of the IPA contradicting the Council of the Association?

Section Letter g currently says In 1948, the Council of the Association recognized ⟨ɡ⟩ and ⟨⟩ as typographic equivalents[...]. [T]he 1949 Principles of the International Phonetic Association recommended the use of ⟨⟩ for a velar plosive and ⟨ɡ⟩ for an advanced one for [some] languages. Isn't that a contradiction? ◅ Sebastian 10:56, 11 January 2022 (UTC)

Conventions change, and the suggested distinction was never widely accepted. Quite a few symbols have changed over the years. Tone diacritics today mean something different than what they did then. In the current Handbook, the velar plosive is illustrated with ⟨ɡ⟩ but afterwards they say, "either of the variant letter shapes ⟨g⟩ or ⟨ɡ⟩ may be used to represent the voiced velar plosive." — kwami (talk) 23:50, 11 January 2022 (UTC)

Yes, change is ubiquitous. But, if I'm parsing the paragraph correctly, this wasn't just commonplace linear change, but circular. If this is really what happened, then the current wording is obfuscating it. It would be clearer if narrated chronologically, along the lines of “In 1948, ... . The next year, this decision was partly revoked when ... . Then in 1993, the original decision was reaffirmedreinstated.” Any objections to changing the text thusly? ◅ Sebastian 05:41, 12 January 2022 (UTC)

I doubt it was 48 to 93. It was a recommended change, but there were lots of recommendations that never went anywhere -- recommended symbols that still aren't in Unicode because hardly anyone ever used them. — kwami (talk) 07:16, 12 January 2022 (UTC)

Yeah, ⟨ɡ⟩ has been the official symbol since 1900. In 1948 the standard 'g' was accepted as equivalent, though it was stated that some might want take advantage of the difference. But essentially no-one ever did. Summary at History_of_the_International_Phonetic_Alphabet#1949_Principles. — kwami (talk) 07:21, 12 January 2022 (UTC)

Thanks, the text in that history section is somewhat clearer.

I now realize that the text here is largely redundant with the history article. How about cutting it down to just a statement like ‘ ⟨ɡ⟩ and ⟨⟩ are approved as typographic equivalents now. See history for details.’? This sentence then could simply become one of the Notes under ‘Letters’ – ‘Consonants’ – ‘Pulmonic consonants’. The current dedicated section right under ‘Design’, on the same level as such fundamentally important concepts as ‘Brackets and transcription delimiters’, bloats this detail out of proportion. ◅ Sebastian 12:35, 12 January 2022 (UTC)

Sounds good to me. — kwami (talk) 21:13, 13 January 2022 (UTC)

No complete Braille IPA support for tones?

I considered adding that based on the detailed article, but decided to wait until I had time to check https://iceb.org/icebipa.htm and https://www.pharmabraille.com/wp-content/uploads/2014/11/World-Braille-Usage-Third-Edition-1.pdf in detail. Comparing section 2.7 (starting on page 24) of the inkprint version of the former (https://iceb.org/IPA-braille_print-ed_final.pdf) and Suprasegmentals and Tones And Word Accents on page XV of the latter with the list in International_Phonetic_Alphabet#Suprasegmentals, I don't see any missing even in the current and original Chao letters, Do others? The Crab Who Played With The Sea (talk) 01:12, 27 February 2022 (UTC)

It only has support for the diacritics and tone letter ligatures that were on the IPA chart at the time, not for others used in the IPA Handbook and not for tone in general.

So, for example, with diacritics it has e᷄ and e᷅ but not e᷇ or e᷆, e᷈ but not e᷉. With Chao tone letters, it has only 4 two-letter ligatures (˩˥, ˥˩, ˧˥, ˩˧) out of the 20 possible (or of the 8 in Chao's original, more restricted set from the 1930s), and only 1 three-letter ligature of the 120 possible (or of the 10 in Chao's original set), and no clear way to extend the system for the rest, or e.g. for Chao's example of a four-letter ligature, [꜔꜒꜖꜔]. So it would not be possible to set the IPA Handbook in Braille without creating additional conventions for tone, such as a sign to indicate that the following letters form a ligature, or just giving up and using ˩˧ for a rising tone but ˧​˩ for the corresponding falling tone.

— kwami (talk) 07:08, 27 February 2022 (UTC)

So perhaps a more accurate statement (here and in the Braille IPA article) would indicate complete support for the IPA chart but not for the IPA handbook (or the additions to it since 2008)? The Crab Who Played With The Sea (talk) 17:42, 27 February 2022 (UTC)

I don't see how that's more accurate, and it isn't particularly relevant. This isn't an article about the chart or the Handbook, but about the IPA, and IPA braille does not fully support the IPA. Also, it doesn't have complete support of the chart today, as one of the sample tone letters has been changed. I think the best we could do is say that tone in IPA braille is "incomplete".

It wouldn't be difficult to make it complete. E.g., of the 3 missing diacritics, 2 have obvious choices, but the 3rd doesn't and a decision would need to be made from the several possibilities. Those patterns would presumably be used for the equivalent tone letters as well. The obvious solution for the thousands of remaining compound tone letters would be to combine the provided tone letters (or maybe just the basic 5) with the ligature mark (point 5), since they are all ligatures. But the established pattern is to not use the ligature mark for tone-letter ligatures. (At least, I haven't seen an example of it.) All we would need is a simple statement by the author or a braille authority. I asked for clarification years ago, but never got a response. If you can find an example of tone letter ligatures (a publication using point 5 between the provided tone letters), then I'd accept that as solved, and the system would be complete if asymmetric. (Though we'd still be missing the 3 diacritics.) — kwami (talk) 02:05, 28 February 2022 (UTC)

You have a point for s/No/Incomplete/ - I may have meant to write "No complete..." and left out a word, or gotten confused; I don't remember. But a word to the wise: I would suggest being very careful, unless you regularly or exclusively write IPA symbols in Braille and read them by touch (do you? I don't, I can just read Braille by sight, or could, and I'm interested in making things as accessible as I know how, can, and should, but learned the hard way not to substitute my judgement to that of people with more directly relevant lived experience than mine) not to say something is obvious and easy. That way lie benevolent ableism and the sighted savior attitude, both of which would be at best very demeaning, and likely also harmful and detrimental to the goal you're trying to achieve. The Crab Who Played With The Sea (talk) 05:19, 28 February 2022 (UTC)

Since for both the diacritics and the letters, Englebretson traced the shapes of Chao's contour tone letters (rising goes up, falling goes down), I expect they'd be the obvious choice for anyone who considers how the system was designed. Though perhaps seemingly arbitrary for those who learn it by rote, but if so most of the system would seem arbitrary. As for the ligatures, Englebretson provided a ligature mark for IPA ligatures, so an obvious way to write the ligatures would be to use the ligature mark. I doubt you need to be sighted to make that connection. — kwami (talk) 06:52, 28 February 2022 (UTC)

Uh... Regarding who is competent and entitled to declare a solution obvious, I think you somehow got the exact opposite of what I wrote. I didn't write anything about needing to be sighted. Instead, I explained why sighted people, including sighted users of Braille, should take a back step and not make any claim of obviousness. What may appear obvious to you (Braille literacy status unknown) or I (halting Braille reader by sight) is not guaranteed to be a good solution, or a solution at all, for someone reading Braille fluently using their fingertips.

Since the IPA handbook in general and what it says about tones specifically are part of why we're having this discussion, how much in it is new since the 1999 edition? (That's where the Braille IPA book got the samples in its appendix from, so I expect it to be at least current to that.)

Last, all the pitch marks you listed as missing are all contour pitches as far as I can tell, and I think all level pitches have one-for-one equivalents. Can you or someone confirm that? The Crab Who Played With The Sea (talk) 11:33, 28 February 2022 (UTC)

Yes, they're all contour pitches. None have changed since the 1999 Handbook. That is, IPA braille could not be used to typeset the 1999 Handbook without extensions for tone.

I hesitate to call them "level pitches", because a lot of IPA use has ligatures for level tones as well, which IPA braille does not support. I think one could argue that the difference is trivial and so could be ignored like the difference between a tie bar and a ligature for affricates is ignored, but I wouldn't make that claim on WP without some reference. But yes, the simple tones (extra low, low, mid, high and extra high for the diacritics; low, mid-low, mid, mid-high, high for the tone letters) are all supported by IPA braille. In addition, three of the six common compound diacritics are supported, as are their tone-letter equivalents. More complex compound diacritics are not supported, but they're exceedingly rare and AFAICT are not mentioned in the Handbook. But the 1999 Handbook specifies other compound tone letters that never made it into IPA braille, at least not overtly. I would assume that was an oversight, that IPA braille was based off either the IPA chart or the appendices of the Handbook (which were based on the chart), and so lacks anything that appeared only in the body of the Handbook, such as the explanatory material or sample illustrations. I suspect it was the chart rather than the appendices of the Handbook, because two of the superscript letters in the appendices -- ˣ and ᵊ -- are missing from IPA braille. — kwami (talk) 23:27, 28 February 2022 (UTC)

(Specifically, IPA braille would need expansion for sect. 2.7 'Suprasegmentals' in the intro; for the Mandarin, Cantonese and Thai examples on p. 24 of the intro; and for the (European) Portuguese illustration in the main text.

(It would also have trouble with the Japanese illustration, since that uses a superscript 'd', and superscript + 'd' in IPA braille is defined as the 'no release' diacritic.) — kwami (talk) 04:02, 1 March 2022 (UTC)

Bracket examples please

For me, the list in § Brackets and transcription delimiters would improve by adding actual examples. That is, when ralistically possible, and in a separate line (not in the description). DePiep (talk) 06:33, 18 August 2022 (UTC)

I made a start. — kwami (talk) 08:13, 18 August 2022 (UTC)

Play Button

There should be an explanation of the round black-and-white icon that looks like a Play Button. It appears everywhere in this article. What does it do? nothing happens when I click it, and there is no sound file associated with it. What am I missing here?

Mang (talk) 15:48, 4 September 2022 (UTC)

I don't see anything that looks like what you describe. Can you give the exact location (section, surrounding text) of one such bytton?1 The Crab Who Played With The Sea (talk) 04:18, 7 September 2022 (UTC)

Hello! Here is an example of five locations of the button: in the third paragraph of the International Phonetic Alphabet article:

"Segments are transcribed by one or more IPA symbols of two basic types: letters and diacritics. For example, the sound of the English letter ⟨t⟩ may be transcribed in IPA with a single letter: [t](BUTTON LOCATION), or with a letter plus diacritics: [t̺ʰ](BUTTON LOCATION), depending on how precise one wishes to be. Slashes are used to signal phonemic transcription; therefore, /t/(BUTTON LOCATION) is more abstract than either [t̺ʰ](BUTTON LOCATION) or [t](BUTTON LOCATION) and might refer to either, depending on the context and language."

There are many more of this button throughout the article. This button also appears many times in this very talk page, mostly in the first half.

Thanks for your help! Mang (talk) 14:46, 9 September 2022 (UTC)

Pinging @Namangwari: — kwami (talk) 08:35, 9 September 2022 (UTC)

Play buttons are in IPA vowel chart with audio, IPA_pulmonic_consonant_chart_with_audio, IPA non-pulmonic consonant chart with audio. Is the issue in there, Namangwari? -DePiep (talk) 08:45, 9 September 2022 (UTC)

Thank you for linking to those three Charts With Audio, DePiep! I see that there are two types of buttons for each sound. While one type is the button that was the subject of my original inquiry, the other type of button is linked to sound files such as File:Close-mid central unrounded vowel.ogg -- while this doesn't answer my question, it does solve my issue of how to hear someone pronouncing these sounds (without resorting to pasting individual letters into http://ipa-reader.xyz/). Mang (talk) 14:57, 9 September 2022 (UTC)

I don't understand "two types of buttons" (I see only one). But it's answered then. BTW, individual sounds like Close-mid front rounded vowel have the same button in their infobox. DePiep (talk) 15:09, 9 September 2022 (UTC)

I assume they're talking about , which is linked to the file description. Nardog (talk) 16:12, 9 September 2022 (UTC)

That still doesn't answer the question as to what "round black-and-white icon that looks like a Play Button" you were talking about. Nardog (talk) 16:12, 9 September 2022 (UTC)

See the quotes by OP @14:46, like more abstract than either [t̺ʰ](BUTTON LOCATION) or .. (in the lede). Actual wikitext: more abstract than either {{IPA|[t̺ʰ]}} or ... That is: an effect of {{IPA}}.

Expanded: [t̺ʰ] ⇆ <span title="Representation in the International Phonetic Alphabet (IPA)" class="IPA" lang="und-Latn-fonipa">[t̺ʰ]</span>. {{IPA}} has no local WP:TEMPLATESTYLES nor -references. This is as far as I know. Does the lang= have extra effects in the UA? -DePiep (talk) 16:43, 9 September 2022 (UTC)

Oh, sorry I missed that. Sounds like Namangwari has installed a browser extension like this. Nardog (talk) 16:56, 9 September 2022 (UTC)

No problem, I had missed it too :-) Thanks for pointing this out. Now, OPs 14:57 post looks like it is solved for them, already (by goiong to our audio tables). This was just an aftertalk. What to do? Looks like the issue is with the Chrome extension. Out of my depth. I propose to leave it. DePiep (talk) 19:24, 9 September 2022 (UTC)

Yeah, it would be the extension's problem, not Wikipedia's. Nardog (talk) 05:53, 10 September 2022 (UTC)

precomposed ⟨ɚ⟩ and ⟨ɝ⟩

§Superscript IPA has the sentence

The precomposed rhotic vowel letters ⟨ɚ ɝ⟩ are not supported, as the rhotic diacritic should be used instead: ⟨ᵊ˞ ᶟ˞⟩; similarly with other rhotic vowels.

⟨ᵊ˞⟩ and ⟨ᶟ˞⟩ seem very strange. Why should ⟨ə⟩ and ⟨ɜ⟩ be written superscript? They are not diacritics, they are the letters to which the rhotic diacritic ⟨˞⟩ is attached.

14:59, 16 September 2022 (UTC) Thnidu (talk) 14:59, 16 September 2022 (UTC)

If you wish to hard-superscript ⟨ɚ ɝ⟩ (examples in the lit can be seen in the Unicode proposal), then you'll need to do it by superscripting ⟨ə ɜ⟩ and then adding the diacritic. Or, if it's not semantically distinct and you don't mind losing the superscripting in a database or file name, you could use soft HTML formatting. The IPA felt no need for Unicode characters for superscript ⟨ɚ ɝ⟩; they judged ⟨ᵊ ᶟ⟩ to be adequate for any conceivable need.

The only reason this is even worth mentioning is that Unicode has the precomposed vowel-diacritic combinations ⟨ɚ ɝ⟩, and since the IPA requested superscript variants of all IPA vowels, a reader might expect that Unicode would therefore have superscript ⟨ᵊ˞ ᶟ˞⟩.

As for why you'd want to do such a thing, it would most likely be for a weak or epenthetic vowel (I saw that once, for rhotic vowels in English, but don't remember where); other possible uses include vowel-coloring of a consonant (e.g. the allophone of /t/ before and after /ɝ/ would be [tᶟ˞], [ᶟ˞t]). — kwami (talk) 19:16, 16 September 2022 (UTC)

Is this an error?

The letters ʎ̝̊, ʟ̝̊, ɭ̥̆, and ɭ̆ have been replaced with letters that are represented with a rectangle with an X. Is it just me? Chiagozie Elobuike (talk) 03:45, 4 February 2023 (UTC)

You need a supporting font. The SIL fonts work well (Gentium, Andika, etc.).

These letters are not on the official chart, but the IPA has requested unicode support for them as implied IPA symbols; the first three are in extIPA, which is acceptable for IPA transcription. — kwami (talk) 06:16, 4 February 2023 (UTC)

Article size

I'd like to put some serious work into improving this article, and one of the things that jumps out at me as something to possibly be improved is the article's size. Not counting bulleted lists, tables, references, or the mountain of footnotes (the tool I use doesn't account for those), the article size still comes out at ~55kB. Per WP:SIZERULE, especially with the addition of the aforementioned uncounted categories, the article could use a good pruning and/or splitting. Not only is the article's text size too large, but it's also made physically harder to navigate by the use of wikitables in several places that replicate the IPA charts. Could we remove some/all of these? I don't know that it's necessary to include any of these charts as wikitables, and if we do, can't we make them much smaller, right-justify them, and use text wrapping like we would for an image? Would it be eurocentric to only keep the pulmonic consonants and vowels? These seem to be the only ones described in detail in (English) intro linguistics textbooks, but obviously those textbooks are written for an English-speaking audience. AviationFreak💬 23:02, 16 February 2023 (UTC)

For an article on the IPA, we want to cover the IPA. We have separate keys for the use of the IPA in English. Charts instead of Wikitables would be fine as long as they were searchable and people could copy from them, but what's the point of doing that? Besides, we currently only have 3 PDF charts and several more tables than that, so we'd need to change all the other tables to searchable charts that respond to the reader's font settings. Is that even possible? — kwami (talk) 06:12, 17 February 2023 (UTC)

Addressing the last point: there are two kinds of introductory linguistic textbooks for English-speaking audiences. 1. Introductions into general linguistics and 2. introductions into the linguistic description of English. We should follow the former. I think there is a silent consensus in WP:LING and WP:LANG that English WP ≠ "Wikipedia about English". –Austronesier (talk) 06:26, 17 February 2023 (UTC)

If they're too much here, we might move the superscript stuff into the Unicode superscript article. I use these charts as a ref all the time, but could just as easily do it from there. — kwami (talk) 06:37, 17 February 2023 (UTC)

Yeah, stuff like the superscripts especially seems like too much for a WP:SUMMARY article on the IPA in my opinion. I agree that English WP coverage of LING topics should not be confined to English phonetics/phonology/etc, but I do think there is a tendency in introductory general ling textbooks (i.e., the former type) to focus on just the vowels and pulmonic consonants (Language Files 12th ed. only talks for a paragraph about other airstream mechanisms and doesn't say anything about their IPA representations). I think this article certainly ought to do more than that, but I don't think it would be inappropriate to talk about the nonpulmonics in a "stronger" summary style. AviationFreak💬 15:25, 17 February 2023 (UTC)

Okay, I moved the charts to the Unicode article, leaving two short paras on the use of superscript IPA. — kwami (talk) 00:27, 18 February 2023 (UTC)

Use of ɒ instead of ɑ for American English

Why does Wikipedia consistently use a symbol for a rounded open back vowel when most Americans do not use that sound? If the answer is that some Americans do, then some accommodation should be made for many regional and other sub-group pronunciations. If the answer is that the rounded sound is standard in England, then some accommodation should be made for Ireland, Scotland, Wales, South Africa, and India, which all have sounds divergent from American or British sounds. 2602:306:C587:B64F:DD36:D2EF:7069:ED3C (talk) 15:11, 30 May 2022 (UTC)

Where is it used for American English? — kwami (talk) 19:10, 30 May 2022 (UTC)

> "Why does Wikipedia consistently use a symbol for a rounded open back vowel when most Americans do not use that sound?" — Because Americans are not the only speakers of English, or users of en-Wikipedia. – _•Raven  ^.talk 02:07, 13 April 2023 (UTC)

Just for the record...

I am Strongly Opposed to any move of this page calling it a "script" rather than an "alphabet", based on the current version of WP:NCWS#Alphabets stating the latter pertains only to "language-specific" alphabets. The discussion on WT:NCWS had made clear: (1) the usage intended was to "one or more languages"; and (2) an alternative qualification was the tiny two words "or use" [-specific]. Neither of these made it into the edit by one sole editor. That omission may not even have been noticed until recently. If the WP page is corrected to express the consensus intention, that issue should vanish. Meanwhile, I cite two RSs:
• Clair, Kate; Busic-Snyder, Cynthia (2012-06-20). "Key Concepts". A Typographic Workbook: A Primer to History, Techniques, and Artistry. Hoboken, New Jersey: Wiley. p. 347. ISBN 9781118399880. alphabet: a set of visual characters or letters in an order fixed by custom. The individual characters represent the sounds of a spoken language. ... In addition to English, there are... International Phonetic Alphabet (IPA),....
• "alphabet". Merriam-Webster [online]. 1.a. a set of letters or other characters with which one or more languages are written especially if arranged in a customary order
– _•Raven  ^.talk 02:27, 13 April 2023 (UTC)

Why are you playing stupid? Do you think that you score points by doing so? — kwami (talk) 09:13, 13 April 2023 (UTC)

Argument by insult and implicit accusation (abusive ad hominem): "This kind of argument, besides usually being fallacious, is also counterproductive, as a proper dialogue is hard to achieve after such an attack." – _•Raven  ^.talk 23:33, 13 April 2023 (UTC)

Has such a move been proposed? —Tamfang (talk) 03:30, 29 April 2023 (UTC)

No, Raven's been making POINTy edits and pretending not to understand anything for weeks. They appear to be spreading their disruption here; it doesn't actually mean anything. They had to be threatened with a block before they'd stop trolling at ANI. — kwami (talk) 04:21, 29 April 2023 (UTC)

and in the absence of any such RtM, Raven's post is unambiguously WP:Disruption. --𝕁𝕄𝔽 (talk) 09:14, 29 April 2023 (UTC)

I would have thought kwami's multiple surprise undiscussed/nonconsensus page moves, which (once reverted as WP:RMUM) kwami re-reverted to keep the changes despite opposition, might have been viewed as disruptive. Since kwami had also declared the IPA to be a "script" and not an "alphabet" (see quotes below), it was entirely possible there would be another WP:RMUM page move based on that declaration or opinion. No point waiting for an RtM, since kwami didn't create any for those other moves. Posting pre–emptive opposition on the talkpage seemed merely prudent. Why kwami now denies making the earlier declarations, ask kwami. – .Raven  ^.talk 03:02, 30 April 2023 (UTC)

> "They had to be threatened with a block before they'd stop trolling at ANI." – Perhaps kwami refers to this exchange:

Please stop. --128.164.177.55 (talk) 12:55, 27 April 2023 (UTC)

When you put it like that... okay. – .Raven .talk 14:01, 27 April 2023 (UTC)

The "trolling" characterization is kwami's own. I for one do not mistake a courteous request for a "threat". – .Raven  ^.talk 03:07, 30 April 2023 (UTC)

See Talk:Theban alphabet#Requested move 3 April 2023 to undo kwami's surprise undiscussed/nonconsensus move (and, after it had been reverted as WP:RMUM, kwami's re-reversion to retain the now clearly opposed move) to 'Theban script', wherein multiple other such undiscussed page moves by kwami were mentioned. Excerpts mention his opinion of the IPA:

Yes, the IPA is a script. In its case, "Alphabet" is an integral part of its name -- we even capitalize it -- as e.g. with the NATO Phonetic "Alphabet", which isn't a writing system at all but instead a set of code words for the letters of the Latin script. There are times when COMMONNAME will override our other conventions, but I'm arguing this shouldn't be one of them. — kwami (talk) 23:44, 3 April 2023 (UTC)

————————————————

To which language do the ISO basic Latin alphabet and International Phonetic Alphabet belong? Do you intend to change those article titles too?

I note that, off-WP, Merriam-Webster defines "alphabet", primary meaning (1a): "a set of letters or other characters with which one or more languages are written especially if arranged in a customary order" – .Raven .talk 04:36, 5 April 2023 (UTC)

They're not alphabets under our convention. 'Alphabet' is just part of the name. Same with the NATO Phonetic Alphabet. Not so for Theban, any more than for Cyrillic or Tengwar. Anyway, your pointed refusal to understand what everyone else is capable of understanding suggests this line of discussion is pointless. — kwami (talk) 04:46, 5 April 2023 (UTC)

[underlines added] Since kwami had made undiscussed (no-request-section) moves of other articles based on such claims, it seemed prudent to put opposition to such a move on record on the talk page. – .Raven  ^.talk 02:53, 30 April 2023 (UTC)