Talk:International Phonetic Alphabet/Archive 10

Help on IPA way of denoting a consonant

Hi, I would like to know the correct IPA symbol to denote a sub-apical retroflex nasal that is pronounced at the palate. That is, it's a retroflex nasal pronounced with the underside of the tongue (sub-apical) pressed against the palate, as opposed to the postalveolar region. I don't know if that would be considered 'palatization', and if so, would ɳ^j be the correct symbol? Thanks. — Jclu (talk) 15:06, 30 January 2009 (UTC)

No, I don't think it would be considered palatalization. I've heard it claimed (no refs off the top of my head but I may be able to find some if you need them) that retroflex sounds cannot be palatalized, or at least are never attested as palatalized. I'd probably use the retraction sign (a small minus sign underneath the symbol) to indicate that the place of articulation is further back than customary. However, even that I'd only do if the distinction between a palatoalveolar and a palatal place of art. is the topic under discussion. If you're just transcribing a language whose retroflex n happens to have a palatal place of art. and there's no discussion of the distinction between that and a palatoalveolar place of art., I'd just use ɳ and explain in prose where the place of art. is. —Angr 15:28, 30 January 2009 (UTC)

Agree with Angr here; I don't think it's articulatorily possible to palatalize a retroflex (except maybe a retroflex dental or something), since that would involve raising both the underside of the tongue and the front of the tongue towards the roof of the mouth...for me, at least, my tongue simply isn't that long or flexible. Politizer ^talk/_contribs 15:36, 30 January 2009 (UTC)

Then is there any way to denote a sub-apical pronounciation? Perhaps I could transcribe the consonant better as a palatal nasal pronounced sub-apically? That should satisfactorily denote the palatal place of art., and include the retroflex component since one would have to curl the tongue back to pronounce with the sub-apical part of the tongue. — Jclu (talk) 16:03, 30 January 2009 (UTC)

Hm...as far as I know, there is no "official" diacritic for sub-apical pronunciation (whereas there are diacritics for apical, laminal, yada yada)...I don't know if there is one that's accepted but just not on the chart, or whatever. May I ask what language this is and what you're transcribing it for? Depending on the project, it might be ok just to make up some convenient notation and include it in your legend/key... Politizer ^talk/_contribs 16:11, 30 January 2009 (UTC)

If the subapicality of this retroflex consonant is not directly relevant to the discussion at hand (i.e. unless you're comparing a subapical allophone with some other allophone, or unless you're comparing a subapical retroflex in this language with some other kind of retroflex in another language), there's no reason to transcribe it as anything other than [ɳ]. IPA symbols are flexible; if only one kind of retroflex is under discussion, you can use the unadorned retroflex symbol for it. —Angr 16:30, 30 January 2009 (UTC)

Thanks for the responses. I'll just use the retroflex nasal symbol. As to the project, it's actually a personal project. Every so often I like to create a conlang; even if nothing comes of it, it gives me a practical opportunity to apply various linguistic theory. — Jclu (talk) 19:22, 30 January 2009 (UTC)

By the way, in The Sounds of the World's Languages, Peter Ladefoged and Ian Maddieson do use separate symbols for apical/postalveolar retroflexes and subapical/palatal retroflexes: for the subapical/palatal ones they use the IPA symbols [ʈ ɖ ɳ] etc., while for the apical/postalveolar ones they use the non-IPA symbols [ṭ ḍ ṇ]. —Angr 11:15, 31 January 2009 (UTC)

"Explicitly recognized?"

Both the page on voiced retroflex implosive and this page state that the Unicode symbol for voiced retroflex implosive is "not explicitly recognized" by the IPA, but is supported in the Unicode Phonetics Extension Supplement. Should the phrasing be "not officially recognized"? Explicit recognition doesn't make sense in this context. If the phrasing should be changed, let me know and I can change it on both pages. Thanks. — Jclu (talk) 20:01, 30 January 2009 (UTC)

Why shouldn't explicit recognition make sense? It means the symbol doesn't actually appear on the chart, but it's fully consistent with IPA principles, so in that sense it's implicitly accepted. —Angr 11:20, 31 January 2009 (UTC)

Well, we've just demonstrated by the event that the phrase "not explicitly recognized" sometimes needs to be explained, and you explained it as not appearing on the IPA chart, so why don't we say "not appearing on the IPA chart"?

• Although not confirmed from any language, and therefore not appearing on the IPA chart, a retroflex implosive, [ᶑ], is supported in the Unicode Phonetic Extensions Supplement, added in version 4.1 of the Unicode Standard, or can be created as a composite [ɗ̡].

Pi zero (talk) 12:15, 31 January 2009 (UTC)

Actually, [ɗ̡] is [ɗ] with an old-style palatalization mark (U+0321). The retroflex hook (U+0322) points to right: [ɗ̢]. — Emil J. 12:15, 2 February 2009 (UTC)

Good catch. Some fonts get them mixed up. kwami (talk) 12:54, 2 February 2009 (UTC)

cover symbols

These aren't part of the IPA, but should be covered somewhere.

C for "any consonant", V for "any vowel", G for "glide", N for "any nasal", F for frics, S for sibs, K for velars, T for alveolars, A for low vowels, R for rhotics, L for laterals, etc. Used for archiphonemes, phonological rules, and reconstructions.

Any ideas where to put them? kwami (talk) 23:06, 6 March 2009 (UTC)

Surely not in the IPA article, as they don't belong here. But I can't think of an article to put them to, either... maybe something general about phonetic transcription? — N-true (talk) 14:20, 7 March 2009 (UTC)

Would it be better to cover them in the articles on what they're used for, such as phonological rule (which does not currently exist, but I might make a stub out of sometime soon)? rʨanaɢ ^talk/_contribs 14:25, 7 March 2009 (UTC)

Phonetic transcription covers all sorts of scripts, and I couldn't think of an on-topic way of working them in. A phon. rule article might be best. Or maybe a "phon. cover symbol" stub to merge into a phon. rule article later on? kwami (talk) 19:59, 7 March 2009 (UTC)

A phon. cover symbol article would probably be nice, if you have enough to write about it. By the way, I created User:Rjanag/Phonological rule to start the phon. rule article; I probably won't have time to expand it until tonight or next week, but my plan is to at least include some examples of rules and have a little diagram or whatnot pointing to what each part of the rule means ("this is what changes...this is what it changes to....this / means 'in the environment where'... this part is the environment..." etc.). rʨanaɢ ^talk/_contribs 20:05, 7 March 2009 (UTC)

Since the cover symbols would be just a list, how about we agree on them here, and you can incorporate them into your article when you're ready?

C: consonant

V: vowel

N: nasal

S: sibilant

K: velar/dorsal

A: low vowel

Etc. There's enough variation from text to text that I don't know how much we want to set in stone. How far should we expand this for maximal utility? kwami (talk) 02:07, 8 March 2009 (UTC)

A few more:

R: sonorant/resonant

X: non-sibilant (maybe not as common)

T: stop

σ: syllable

I would add to either phonetic transcription (although it's not phonetic) or natural class (although not all classes have nice abbreviatory symbols). This is really about phonological representation not phono rules (you can use the same symbols in constraints). – ishwar  (speak) 02:58, 8 March 2009 (UTC)

If we're listing things that aren't purely featural, like σ, then we could also list # , for word boundaries. Who knows what else is out there, I'm not actually a phonologist, I just lurk around here :) rʨanaɢ ^talk/_contribs 03:25, 8 March 2009 (UTC)

Contradiction?

"Among the symbols of the IPA, 107 represent consonants and vowels, 31 are diacritics that are used to further specify these sounds, and 19 are used to indicate such qualities as length, tone, stress, and intonation." (Section 2, "Description") "As of 2008, there are 107 distinct letters, 52 diacritics, and 4 prosody marks in the IPA proper." (Introduction) Is this a contradiction? —Jchthys 16:00, 12 March 2009 (UTC)

Yeah, we seem to be off by 6. kwami (talk) 19:28, 12 March 2009 (UTC)

THE BIGGEST CONTRADICTION is the international alphabet is recognized as: ALPHA, BRAVO, CHARLIE, etc. all this bs is NOT the recognized "international alphabet" —Preceding unsigned comment added by 65.101.235.167 (talk) 14:31, 27 May 2009 (UTC)

The "Alpha, Bravo, Charlie" system is neither international, nor phonetic, nor an alphabet, so it can hardly be called an international phonetic alphabet (let alone the International Phonetic Alphabet). +Angr 14:44, 27 May 2009 (UTC)

There's already a hatnote on this article, {{distinguish|NATO phonetic alphabet}}. —Pi zero (talk) 16:05, 27 May 2009 (UTC)

transcribe this article using the IPA

Wouldn't it be more informative to transcribe parts of this article using IPA? —Preceding unsigned comment added by 69.250.186.142 (talk) 13:32, 22 March 2009 (UTC)

Parts of the article, no. A single example sentence or two, maybe. rʨanaɢ ^talk/_contribs 13:40, 22 March 2009 (UTC)

Link to ipatrainer.com

Hi! I'm just wondering why the link to ipatrainer.com was removed? I feel it is a very relevant page.

--129.242.182.27 (talk) 07:18, 31 March 2009 (UTC)

Please see WP:EL for Wikipedia practices on the inclusion of external links. —Angr 09:29, 31 March 2009 (UTC)

GA Sweeps

This article has been reviewed as part of Wikipedia:WikiProject Good articles/Project quality task force. I believe the article currently meets the criteria and should remain listed as a Good article. The article history has been updated to reflect this review.

I passed it under some doubt; though it's generally speaking a very good article, it is somewhat low on references, and would probably not have passed if it was nominated today, without some work. The references need some cleaning up, e.g. the first one has no page number. The article also include some external links in the text, which should be avoided. Lampman (talk) 14:43, 13 June 2009 (UTC)

What is the source of the egressive consonant table and its POA categories

What, exactly, is the source of the IPA egressive consonant chart used in this article? It differs from the official IPA chart in several ways, notably in the place of articulation categories (labial, coronoal, dorsal, radical), which are *not* in the official IPA chart. Where did these come from? Are they reliable? Why are they here at all? —Preceding unsigned comment added by 190.20.201.145 (talk) 07:14, 15 August 2009 (UTC)

It's a bit more user friendly, that's all. Both this and the chart found in the IPA Handbook are summaries, not the complete IPA. As for coronal, etc., those higher-level categories can be found (among other sources) in SOWL, co-authored by a former president of the IPA. kwami (talk) 12:48, 15 August 2009 (UTC)

Why doesn't somebody address the obvious???

Threads moved to Wikipedia talk:IPA#Why doesn't somebody address the obvious??? and Wikipedia talk:IPA#Why doesn't somebody address the obvious? (redux)

External Links

I added three new links to External Links to IPA charts to help people who have trouble learning it. Some of them play sound samples of each phoneme and two of them require Flash. I hope this is helpful. Mossman fmcb (talk) 10:07, 12 September 2009 (UTC)

Are all of these necessary? And, more importantly, would you be able to integrate these links into the subsections that already exist, rather than creating a new subsection where there are already too many?

For example, off the top of my head I can think that Peter Ladefoged's website already has sound samples of phonemes, like the links you added, and is already included in a subsection. rʨanaɢ ^talk/_contribs 00:20, 17 September 2009 (UTC)

I put those links there so that those who are unfamiliar with the IPA could go there and easily learn how it works and what the symbols actually represent. There may be a lot of people out there who don't understand the IPA enough to be able to read and understand phonetic transcriptions, so I just thought it might be a little helpful. Mossman fmcb (talk) 02:34, 12 December 2009 (UTC)

May I suggest an external link to Voices.com's article on the International Phonetic Alphabet that discusses how to read an IPA chart and the three chart categories: pulmonic consonants, non-pulmonic consonants, and vowels. From my research, this page has been linked to from the linguistics department at Berkeley, The Illinois Institute of Technology and others. Thank you for your consideration. Davidciccarelli (talk) 02:55, 20 July 2010 (UTC)

Using the Rare [ɡ] Character

• I wonder, why is this character used, when the regular g appears exactly the same (or almost the same) in most fonts. --Mahmudmasri (talk) 20:32, 8 October 2009 (UTC)
  • Most sans-serif fonts, yes, but not in most serif fonts. +Angr 20:44, 8 October 2009 (UTC)

Use in education

I started a small section about the usage of IPA in foreign language education. I only have one example substantial example: IPA was used in Soviet school textbooks for French and English. This does not require a particular citation, because all Soviet schools used exactly the same textbooks. Unfortunately, i don't know whether it was used for other languages or whether it is still used in Russia.

It would be very nice if someone could add more example from other countries. I saw, for example, textbooks for studying Catalan, Portuguese and French with IPA, but these were textbooks for courses for adults; for this matter i personally find children's education more interesting. --Amir E. Aharoni (talk) 15:11, 15 November 2009 (UTC)

This is a useful section. However, you may be missing the point about citations: we still need a citation to support the claim that the IPA was used in Soviet textbooks for schools, regardless of whether all the schools used the same ones. garik (talk) 16:33, 15 November 2009 (UTC)

The book itself is a citation. Soviet textbooks didn't have fancy names: "English, 2nd grade" - that's how all Soviet textbooks were called (a good thing, if you ask me). I studied English from it in 1987 - i don't know about other years. --Amir E. Aharoni (talk) 17:28, 15 November 2009 (UTC)

It doesn't matter as much for this sort of claim, which I suppose is relatively uncontroverisal, as it would for some other claims; however, you're still missing the point. You're essentially making two claims:

1) That all schools in the Soviet Union used, for a certain period (what period?), particular textbooks for English and French;

2) That these textbooks used the IPA to indicate pronunciations.

And you need to support both claims by providing reliable sources, and by referencing them in the article. The book itself may work as a citation, but you haven't referenced it in the article; you haven't even referenced it fully on this page: who wrote it? Where was it published? Which edition are you referring to? The other problem with using this sort of source as support is that it's not very easily available to the average reader. If you have a link to an online copy of the book, this would be very helpful. As I say, this is a useful section, and I don't doubt your word, but you need to read up on citing sources. The only reason why this claim might not need a citation is that people may be unlikely to challenge it; it has nothing to do with the fact that all schools used the same textbooks. garik (talk) 19:18, 15 November 2009 (UTC)

Sorry, i don't remember the author's name. Maybe it was the 1987 edition of this book by Irina Vereschagina, but i'm not 100% sure. --Amir E. Aharoni (talk) 20:03, 15 November 2009 (UTC)

OK, does anyone have any other sources to back up the claim? I'm pretty sure it's probably accurate; my own experience of teaching in Russia suggests that the average Russian school pupil even today is considerably more familiar with the IPA than school pupils in, say, Western Europe. However, it would be good to have a source. Apart from anything else, there are a lot of alphabets around used to represent pronunciation, and it would be easy for someone, from whatever country, to think that their schoolbook used the IPA, when in fact it had its own ad hoc transcription system. So it would be good not to set a precedent. garik (talk) 11:29, 16 November 2009 (UTC)

I asked a few people to help me find a real school text book, but i am 100% sure that it was definitely the standard IPA in my school text book, with æ, ə etc. At home i have several Soviet English-Russian dictionaries and an English textbook for adults (L. V. Bankevich, "English Through Eye and Ear", Vysshaya Shkola, 1975) and they all use the standard IPA. There's no reason that school textbooks would use anything else. --Amir E. Aharoni (talk) 19:10, 16 November 2009 (UTC)

And yet all sorts of textbooks do use all sorts of alternative ad hoc transcription systems. I believe you that the Russian textbooks used the IPA, but Wikipedia is about verifiability NOT truth. As the page states: what matters is "whether readers are able to check that material added to Wikipedia has already been published by a reliable source, not whether we think it is true". In other words, even if you wrote the textbook, your word on this page would not be sufficient support for the claim to be included if challenged. To support a claim made on Wikipedia, we need a reliable published source, ideally in English (for English Wikipedia) that readers are likely to be able to check. To put this in perspective: if Barack Obama edited the Barack Obama article, he would still need to back up his edits with reliable published sources. His own word (unless published in such a reliable source) would not be good enough. garik (talk) 19:22, 16 November 2009 (UTC)

I came here looking for the rounded-w 'oo' sound from my first-year primary (1971 Essex/UK) school books. It was actually ITA (Initial Teaching Alphabet). Not sure whether it's relevant as a 'See Also' in this topic. SeanCollins (talk) 18:36, 17 November 2009 (UTC)

I don't think links to individual phonetic alphabets should be included. There are hundreds, if not thousands, of them, and there's no space to list them all, so there's no point listing random ones. rʨanaɢ ^talk/_contribs 18:44, 17 November 2009 (UTC)

Isn't this a no-brainer?

Regarding this and similar edits: I haven't commented on t his topic before, but I'm beginning to think this whole topic (of schools using IPA to teach language) is unencyclopedic, or at least redundant. IPA is a phonetic notation system, indicating pronunciation is exactly what it's for, so of course it's used in some educational settings. (To be honest, the only thing surprising is how little it's used in education, for example in the US.) Is this realy worth writing a whole section about?? rʨanaɢ ^talk/_contribs 10:36, 21 November 2009 (UTC)

Actually, that is a good point. It may be worth replacing it with a section headed "use in schools", where we include some stats (if they exist) about what proportion of countries teach the IPA to school kids, or at least use it to teach them foreign languages. That "if they exist" is a big if though. garik (talk) 10:56, 21 November 2009 (UTC)

That's exactly the point: It may make sense to some people that it ought to be used in schools, but it is not used everywhere. I'd love to know where it is used and where not and i don't understand how it's different from the encyclopedicity of usage in dictionaries or in linguistic research. If it's used in schools, at least in some places, it means that there it is not a weird alphabet for professors, but an educational tool for everyone.

We grownups rarely think about school textbooks in our own countries, let alone abroad. I shared a piece of knowledge that i have - that it's used in Russia. AFAIK it is not used in schools in Israel, where i live now, but i don't put it in the article, since i don't have hard data (i'm looking for it).

The many arguments about the pros and cons of IPA's usage in Wikipedia may well be related to its not being used in schools and dictionaries in countries where English is native. --Amir E. Aharoni (talk) 11:17, 21 November 2009 (UTC)

That's probably true. I do think that this section should be about the big picture, if at all possible. The particular example of Russia (and the former USSR) doesn't tell us very much on its own. We need to know if Russia is exceptional and, if so, how exceptional. So good luck with your search for data! I'll try to have a look myself. garik (talk) 12:18, 21 November 2009 (UTC)

I totally agree. That's why i put {{Expand section}} on it. I think that it is better than nothing. --Amir E. Aharoni (talk) 12:55, 21 November 2009 (UTC)

I just hope that we can actually expand the section with some big-picture data. If, instead, all we get is a couple of examples of the IPA being used in school books, then the section is indeed probably better than nothing, but not by much! garik (talk) 14:22, 22 November 2009 (UTC)

Math Brackets for Linguistics?

I don't think it's a good idea to use math brackets here, as in ⟨foo⟩. It's not a common practice either.

• The Unicode chart explicitly says that U+27E8, U+27E9 are mathematical bracket, that is so-called bra and ket in math. What we need is not exactly math symbols: notation like <foo> is for clarity, simply <foo> with ASCII < and > is ok too.
• The Unicode Standard explicitly says, both single and double guillements may be treated more like brackets than quotation marks. So we can use ‹foo›—if for some reason we don't like < and > then ‹ and › are handy because many fonts support them.
• Charis SIL and Doulos SIL do not support U+27E8/9, suggesting that they are not supposed to be commonly used by linguists. U+27E8/9 are supported by only a few fonts (Code2000, Cambria Math, DejaVu) and as such, maybe average users cannot see these symbols anyway. If we really need those characters, perhaps we should use inline images (as in bra-ket notation). But I don't think so. Actually <foo> is just fine, as everyone was happy for a long time with it. If < and > look “unprofessional”, how about ‹ and › ? :)

Using those math symbols is technically not wrong, it's an ok option, but imho not very advantageous. —Gyopi (talk) 08:04, 23 November 2009 (UTC)

AFAIK, the math symbols are the correct ones. (After all, Unicode doesn't have separate math & non-math square brackets, braces, or parentheses; it's just that in this case mathematical use is more common than the epigraphic one.) At least, they are the only angle brackets supported by Unicode outside of CJK stuff, which causes spacing as well as accessibility problems.

The angle-bracket notation is used more by epigraphers than by linguists, which might explain why SIL didn't cover them. Both <...> and guillemots are common approximations, though true angle brackets tend to be more common in pubs with high quality typesetting.

The real problem, as you point out, is lack of font support. I suppose we should go back to guillemots. The problem with <...>, beside them being ugly IMO, is that they are ubiquitous in WP coding, making any future conversion a real pain in the ass. If we gradually change the articles over to guillemots as we're editing them for other things, then it will be really easy to semi-automatically switch them over to true angle brackets if and when IPA fonts start to support them. kwami (talk) 08:40, 23 November 2009 (UTC)

Ƶ§œš¹ tells me the angle brackets ⟨ ⟩ don't show up on his computer. Does anyone else have that problem? (Here they are with ⟨ ⟩ and without ⟨ ⟩ IPA formatting; Unicode: ⟨ ⟩.) Probably if he can't read them, there are many others who can't. kwami (talk) 09:16, 23 November 2009 (UTC)

Shows up neatly for me (Win XP / Firefox). What OS is Ƶ§œš¹ using? — Sebastian 21:26, 23 November 2009 (UTC)

Also Wikipedia:IPA says: “<Angle brackets> may be used to represent”... We shouldn't change this style without a consensus. I can see your point: it's inconvenient that you have to type &lt;i&gt; when you want to show <i> but even so, typing < and > (OR &lt; and &gt;) is much easier for ordinary editors than to input things like ⟨ 〈 〈 ‹ ❬ or ❮. If we really need it, we can easily make a template like {{anglebra|foo}}, though...—Gyopi (talk) 09:45, 23 November 2009 (UTC)

I can't see the proper angle brackets either. What's wrong with keeping the normal keyboard < and >?−Woodstone (talk) 10:12, 23 November 2009 (UTC)

They're sloppy, and are a bitch to edit. Guillemots at least look professional, even if they're the wrong height, and they're so little used on WP-en that they'll be easy to convert in the future. kwami (talk) 10:25, 23 November 2009 (UTC)

I support the idea of using a template like anglebra. That way, we can separate the layout decision from the way we enter the text, and we can start to gradually change the articles, as Kwami proposed. — Sebastian 21:26, 23 November 2009 (UTC)

Heard back from SIL. Their fonts currently support U+2329 and U+232A for the angle brackets. These are now deprecated by Unicode "for mathematical use because of their canonical equivalence to CJK punctuation" (not an issue, I think, for us), but they work in my browser and don't cause the spacing issues of the actual CJK punctuation (U+3008/9). So, do these look good?

displaying angle brackets U+2329 and U+232A

unformatted	{{IPA}} formatted	{{Unicode}} formatted
〈a〉	〈a〉	〈a〉

As for a template, we could just add these to the IPA input window when we edit, so they'd be no more difficult than stress marks.

(BTW, I just added the missing IPA diacritics to that window. I've never done that before, so please tell me if there are any problems.) —kwami (talk) 02:16, 24 November 2009 (UTC)

Sorry, somehow although I changed the brackets in this article w AWB, the save didn't go through. So I just changed from math to standard angle brackets; having them on the page might bring out any problems.

Weird: if you change to angle brackets in the edit window, WP changes them to CJK when you save. So in order to get them, we need to use the &#x...; coding, which is maybe the source of some of the confusion above. Wiktionary has the same problem, or at least I have the same problem with Wict. Maybe it's my browser (FF)? Can anyone else enter brakets and have them save properly? kwami (talk) 06:45, 24 November 2009 (UTC)

I think it's Unicode normalization. Normalize( U+2329 ) = U+3008. It appears that the software here automatically does Normalization C. While we as editors can ignore the "Canonical equivalence", Wikipedia's software respects that behind our backs. That's a problem. You can use &#... as a workaround but then other editors cannot copy-paste the displayed brackets. —Gyopi (talk) 07:16, 24 November 2009 (UTC)

The only reason you like U+2329 better than U+3008 is because it's not “wide” in your font. Technically speaking, that's not correct. Check UAX11 and UCD: both U+2329 and U+3008 are in the class W (Wide). Even if some fonts don't treat U+2329 as W, that is not guaranteed in the Unicode standard. It's OK (actually correct) to use a full-width glyph for U+2329. I do like SIL's U+2329 glyph, but this is not a way to go. There are several problems here (obsolete code points, normalization-unsafe, Width problem).—Gyopi (talk) 07:56, 24 November 2009 (UTC)

PS: Historically SIL's implementation was not wrong, if it dates back to Unicode 3.1 or before. In Unicode 3.1 and before, U+2329 was N (Neutral)[1] or A (Ambiguous)[2]: it's W (always wide) in Unicode 3.2 and later. —Gyopi (talk) 08:08, 24 November 2009 (UTC)

Good to know.

Actually, the most serious problem would be lack of font support. Few of our readers are going to have CJK support in their browsers.

But if Unicode has obsoleted these code points, that's a real problem. kwami (talk) 08:11, 24 November 2009 (UTC)

Currently, IE7 shows a lot of "boxes" (.notdef glyphs) in my tests (with Internet Explorer Collection) for U+2329/U+232A you put in International Phonetic Alphabet. U+2329 (or U+3008) is not good for several reasons above, plus in general people just can't see them right. U+2039 ‹foo› is possible, but the glyphs ‹ and › are usually smaller than we hope... Personally, I like simple <foo> better than ‹foo›. How about just using < and > to make old browsers happy? That they look sloppy is subjective and not a reason enough to confuse readers without a newest browser. In this case, < and > work just as fine as special Unicode characters. Your second reason ("use something different so that you can easily replace them later") is understandable but not logical; essentially you're saying "We should replace them so that we can replace them later." which is a tautology, not explaining "why to replace" (but I do understand. The true reason is < and > look cheap and unprofessional). Even if you believe that someday it will become a common practice to use the math brackets (U+27E8/9) for this purpose, you should replace the characters only when that someday comes. I mean, it may be ok to use the ‹foo› style when you write something new, but I don't think we should replace the existing < and > with ‹ and ›, when the latter is not really good anyway, unless there is a consensus that we use ‹ and › for now.—Gyopi (talk) 09:20, 24 November 2009 (UTC)

They look just fine in IE8. (I don't have IE7 installed.) You're right, none of the three sets of true angle brackets appear to be tenable for the time being.

Yesterday I found over 50 articles that already used ‹...›, and I know there are others. So there is s.t. of a consensus, or at least common WP practice. If we go back to <...>, we'll have lost that work. Now that the angle brackets supported by SIL are deprecated by Unicode, it's only a matter of time before SIL switches over; give extra time for people to update their fonts, and we can follow suit. ‹...› will be easier to convert. Or we could just wait until people update to IE8, if we can work around the copy & past issue. (We might be able to figure out how to add them into the edit window.) kwami (talk) 09:35, 24 November 2009 (UTC)

You can use (1) ‹...› instead of angle brackets. By the same logic, there is an implicit consensus that one can use (2) ASCII <...> instead of them (you'd find more than 50 articles that have <...> already). Both should be equally okay. You don't need to change (1) to (2); you don't need to change (2) to (1) either. Though changing (2) to (1) wouldn't hurt (you don't need to revert, either), some editors might actually like (1) better.

About the original question, you were correct after all :) If you read UAX#11 Modifications carefully, it says (in short) it's correct to use mathematical angle brackets as common angle brackets in English text. The only problem is, they are not well-supported yet. Somewhat like using “ ” ’ ‐ … is better than using ASCII " " ' - ... but not always practical yet. Code 2000, DejaVu, Everson Mono supports those brackets (⟨...⟩). Cambria Math has them too, so Vista users should be able to see them (I'm on Windows XP & not 100% sure though). Btw if you'd like to test MSIE5/6/7 etc, you can install Internet Explorer Collection. I used it when I reported about IE7. —Gyopi (talk) 08:53, 25 November 2009 (UTC) 

Hey, thanks for that. Now we know what it should be; hopefully the next release of Gentium in 2010 will be updated appropriately. kwami (talk) 20:14, 25 November 2009 (UTC)

Yep, you can replace <...> directly with ⟨...⟩ when the SIL fonts are ready. Don't waste your time with tentative replacements with equally bad ‹...› or obsolete U+2329/A 〈...〉. If the article is not technical and for general readers, <...> may be safer and preferable, but if the article is technical, people who read it are likely to have SIL fonts installed.—Gyopi (talk) 08:35, 26 November 2009 (UTC)

‹...› can be quickly swapped out using AWB. <...> cannot. So where we have ‹...›, we wouldn't want to go backwards, and if we gradually convert to ‹...›, the eventual transition will be much easier. kwami (talk) 09:00, 26 November 2009 (UTC)

Sure, don't bother reverting the changes. I wonder what others think about this. Woodstone likes <...>; Sebastian supports a template, rather than direct text replacement. Does anyone have any other opinions? —Gyopi (talk) 11:26, 27 November 2009 (UTC)

I wonder if we could create a template that would prioritize the math brackets if one of the proper fonts are available, and display something else if they are not. kwami (talk) 11:38, 27 November 2009 (UTC)

Good idea! I asked about that at WikiProject Templates. — Sebastian 16:55, 2 December 2009 (UTC)

Just to follow up: The answer was "it wouldn't be worth the effort". — Sebastian 23:28, 10 February 2010 (UTC)

Use in education, part 2

I emailed the International Phonetic Association about this (the "secretary" email that appears on the IPA website) and here's what i got in reply (the person writing is British):

I am not aware of IPA symbols being used in any systematic way in languages education in the UK, although individual teachers do undoubtedly use them as part of their own personal way of teaching.

Where they are used is in the A-Level English Language syllabus. They will, I imagine, also be used in the new Diploma in Languages that is to be trialled soon (an alternative to A-Levels).

They may also start to be used more systematically in the very near future because they will be part of the in-service phonetics training course for primary school teachers who will be teaching a modern foreign language at primary level as part of the new government initiative Links into Languages.

Otherwise, I think - very disappointingly - the answer is that the symbols are not used in any systematic way, in spite of the fact that FL translating dictionaries are increasingly transcribing the pronunciation using IPA.

I am not familiar with the British education system. If anyone is familiar with it, it would be nice if they could put it into the article with proper references to British textbooks.

Thanks in advance. --Amir E. Aharoni (talk) 13:24, 2 December 2009 (UTC)

The term "Symbol" is way overused in this article

Whether or not the IPA calls these tokens symbols from time to time, in many contexts in this article either "letter" or "character" or "glyph" is meant. Where there is precise and more modern terminology available it should be used. Someone just reverted an effort I was going to begin to improve this article. Therefore I have brought the question here. Must we use symbol symbol symbol symbol symbol symbol symbol symbol so much? -- Evertype·✆ 18:25, 29 December 2009 (UTC)

Thanks for bringing this here. You apparently want to increase diversity in the article, as can be seen from your version, which had 115 occurrences of "symbol", and 77 of "letter" (compare with 121 and 71 in the current version). There is a common perception that repeated words are monotonous and should therefore avoided. One of the tricks to avoid it is to replace the word with synonyms.

BTW, this trick is used with more synonyms: We also have the term "character", used synonymously to "letter", as in "Each character is assigned a number, to prevent confusion between similar letters". In addition, the term "marking" is used as a synonym to "symbol": "Diacritics are small markings which are placed around the IPA letter ...".

Introducing variety through synonyms works well for prose writing, but is actually discouraged in technical writing, because it decreases clarity. This might not be immediately obvious, especially for someone who already knows the topic. But just look at the sentence containing "character" above - I am sure that it will take the average reader a little longer to understand than if it simply used "letter" twice.

That all said, one still might object that it doesn't hurt to vary a bit between synonyms when there is no meaningful distinction between them. But is that so?

When you look at the current article, the term "letter" is used for several specific meanings distinct from the meaning of "IPA symbol":

1. Distinction from letters in other writing systems:
   • "The general principle of the IPA is to provide one symbol for each distinctive sound (or speech segment).[10] This means that it does not use letter combinations to represent single sounds,[note 2] or single letters to represent multiple sounds (the way ‹x› represents [ks] or [ɡz] in English)."
   • "... the respelling systems in many American dictionaries (such as Merriam–Webster) use ... only letters of the English Roman alphabet and variations of them."
   • "The traditional names of the Latin and Greek letters are usually used for unmodified symbols."
2. Distinction from other IPA symbols:
   • "It is customary to use simpler letters, without a lot of diacritics, in phonemic transcriptions." (In this example, the usage of "letter" coincides with the usage of "‹›".)
   • "Tone diacritics and tone letters"
   • "... the diacritic would need to be written below the letter."
3. Letter form:
   • "An example of capital letter forms for IPA symbols is Kabiyé of northern Togo, which has Ɔ Ɛ ..."

While those distinctions admittedly are not carried through consistently, I still see them as more helpful to our readers than striving for the ideal of avoiding word repetitions. — Sebastian 20:55, 29 December 2009 (UTC)

IPA, AFI and API

I propose add rhe names used in others languages for to refer to International Phonetic Alphabet. --Der Künstler (talk) 00:45, 9 January 2010 (UTC)

This is English Wikipedia. Thank you, come again. rʨanaɢ ^talk/_contribs 01:25, 9 January 2010 (UTC)

Actually, now that I think of it, you are right. We should include a translation of "IPA" in every language (who cares that they're already in the toolbar, we need them at the top of the article too!). I propose that we add the extremely widely-used abbreviation GJYB and all other translations as well. Anyone who opposes this is suppressing the truth that the world NEEDS to know!!!!!11~ rʨanaɢ ^talk/_contribs 01:29, 9 January 2010 (UTC)

Sarcasm not especially useful. -- Quiddity (talk) 02:35, 9 January 2010 (UTC)

The translations/variants are already given in the sidebar-interwiki links. They wouldn't aid in understanding the topic, and are not incoming redirects. -- Quiddity (talk) 02:35, 9 January 2010 (UTC)

Oppose per common sense. Though I find rʨanaɢ's response to be unnecessarily dismissive. -- Evertype·✆ 09:44, 9 January 2010 (UTC)

Cf. AFI and API. -- Evertype·✆

I agree with Evertype. There is absolutely no need to add abbreviations in other languages; they would contribute nothing more than pointless clutter. And I also think Rjanag's response is unnecessarily sarcastic, though I understand his frustration after the round of silly back and forth reverting that's been happening on the page. garik (talk) 10:36, 9 January 2010 (UTC)

Regardless of whether or not I was too mean, I was also raising a practical point that may have been missed. Assuming we do add extra abbreviations, where would we draw the line? Why just the European API and AFI, why not translations for every language? GJYB is not much more far-fetched than API (although, to be fair, I think a lot of Chinese texts just say "IPA" anyway). rʨanaɢ ^talk/_contribs 15:10, 9 January 2010 (UTC)

In the article, in other wikipedias, is the alternative names used in others languages; AFI (in Spanish, Portuguese, Italian) and API (in French). To say that is not necessary to add the name used in other languages, because this is the wikipedia in English is as to say, is not necessary to add the name Andalucía, in the article Andalusia, because this is the wikipedia in English --Der Künstler (talk) 00:30, 10 January 2010 (UTC)

I think you're missing the point. We give the Spanish name for Andalusia because that's the main language spoken in Andalusia; we don't give the name for Andalusia in French, German, Welsh, Quechua, or any other languages. Moreover, giving a local name is a convention essentially restricted to geographical entities; we don't, for example, list foreign-language names for the United Nations. The International Phonetic Alphabet doesn't even have a local language; the concept makes no sense. If we listed foreign-language names for entities like this, we'd have to list hundreds of names. There's a good reason we don't do it. garik (talk) 14:45, 10 January 2010 (UTC)

[tˢæːld̥ɘ]

What abour the diacritics contained in this word (the ‹s› subscribed adjacent to the ‹t› and the little circle under the ‹d› which sometimes also comes above a letter (Danish voiced velar plosive))? I couldn't find them here in this article. Wisapi (talk) 22:50, 19 January 2010 (UTC)

The circle under <d> represents voicelessness and is included in the chart under "phonation diacritics". The superscript <s> probably represents coarticulation (see "coarticulation diacritics"). rʨanaɢ ^talk/_contribs 22:55, 19 January 2010 (UTC)

Superscripted t^s is mentioned on the IPA chart with explanation "fricative release". — Emil J. 10:56, 21 January 2010 (UTC)

Really? That sounds a lot like an affricate [ʦ]. rʨanaɢ ^talk/_contribs 11:01, 21 January 2010 (UTC)

I believe that superscript s was withdrawn in 1989, possibly because of this redundnacy. As I understand it, it was supposed to represent a very brief voiceless alveolar fricative; so, following a [t], it wouldn't be usefully distinguishable from the affricate [ʦ]. This may be why it was withdrawn. garik (talk) 11:11, 21 January 2010 (UTC)

You are right. I was looking at File:IPA chart 2005 png.svg, and I didn't realize it is not a copy of the original chart. This strongly suggests that the file should be fixed. — Emil J. 11:55, 21 January 2010 (UTC)

[ʦ] is not IPA.

Any IPA letter can be superscripted as a diacritic to add a nuance to the base letter. The raised s is simply one of the more common. But there are raised nasals, raised schwas, etc. etc. kwami (talk) 12:06, 21 January 2010 (UTC)

[ʦ] = [t͡s] = [ts]. Just depends on your conventions. ([ts] is the easiest to type, [t͡s] with the tie bar seems to be the most popular, and unified [ʦ] sometimes pops up when someone wants to only use single-glyph characters in their transcriptions, for computational reasons.) rʨanaɢ ^talk/_contribs 16:35, 21 January 2010 (UTC)

Now I found that little circle; it is so tiny that I thought it was a dot. If I'm not wrong, that may come under or over a letter, right? So we could add <ɡ̊> or <ŋ̊> besides <t̥> and <d̥> to clarify that and to make sure that people percive that it's a ring (the o'er-the-letter version is bigger).

However, I couldn't find such a page as "coarticulation diacritics" nor any subtopic with this name in this article, and in my humble opinion it's impossible that <tˢ> be a coarticulated phoneme. Both <t> and <s> are realised in the place, so the tongue can't block the airway and cause turbulance there at the same time. Wisapi (talk) 20:19, 22 January 2010 (UTC)

Search the page for "Co-articulation diacritics", you will find it (sorry I missed the hyphen last time). As for what it actually represents, you are probably right that coarticulation would be weird...and, for reasons mentioned above, using it to represent affrication would also be weird. This probably explains why I (and some others here) never learned it; seems like it might be deprecated. rʨanaɢ ^talk/_contribs 20:43, 22 January 2010 (UTC)

Hm, now I found it. And I also came upon these arguments supporting your claim. But I'm still confused about it. We should ask a Dane to record that soud comparing it to [t] and [t͡s] and give some explanation. Wisapi (talk) 00:42, 23 January 2010 (UTC)

<d̥> for <t> and <t̬> for <d>

Why do we use voiceless marks on voiced symbols that already have a voiceless counterpart and vice versa (e.g. <d̥> for <t> and <t̬> for <d>)? Wisapi (talk) 00:29, 24 January 2010 (UTC)

Depends. Some languages (incl. many dialects of German) contrast a lenis /d̥/ with a fortis /t/. Both are unvoiced. In some treatments, <t̬> might represent an underlying //t// that is phonemically voiced, whereas <d> would be an underlying //d//. Either <d̥> or <t̬> could also represent a partially voiced [t]. kwami (talk) 00:43, 24 January 2010 (UTC)

Hatnotes

My 2 agorot about the hatnotes war that's going on here lately: One link to IPA (disambiguation) seems enough to me as far as links to article space go. Nothing there seems to me anywhere near the International Phonetic Alphabet in terms of notability. Being a linguist, however, i may be biased, so other opinions are welcome.

I don't see a big problem with having a link to Wikipedia:IPA, though.

Finally, i don't understand at all how can the IPA be easily confused with the NATO phonetic alphabet. --Amir E. Aharoni (talk) 13:01, 27 January 2010 (UTC)

Indeed, strange to me. Three hatnotes! Only the link to disambiguation should stay. The others (ICAO spelling and WP:IPA -- yes, even a wiki-internal!) can go to the See Also section. -DePiep (talk) 22:20, 12 June 2010 (UTC)

Allow me being bold. -DePiep (talk) 21:38, 14 June 2010 (UTC)

Violation of NPA

You people are idiots.

Quote from above: "I learned IPA in less than a week when I was an undergrad!"

Is "less than a week" really an acceptable timeframe for studying a Wikipedia article? People come here to see "what the hell that news lady was talking about," or to find out "was Joe telling the truth at lunch?"

Jesus Christ, people--when someone comes upon an article for a word they can't pronounce, at least one of the pronunciations given should actually help them--not drag them on a wild goose chase for enlightenment where they will invariably become frustrated and give up. The fact that this page repeatedly attracts criticism of this sort should tell you that this line of thought is pretty common.

And then you jerks have the audacity to accuse the offended reader of some kind of imperialism. It's a good thing Wikipedia is free; the customer service sucks.

J.M. Archer (talk) 15:35, 5 March 2010 (UTC)

This talk page is for improving the article, not debating Wikipedia policy. But, if you do take this to the relevant place (Wikipedia_talk:Manual_of_Style_(pronunciation)), I recommend reading the notice at the top of that page. Please also read WP:civil. garik (talk) 15:42, 5 March 2010 (UTC)

Thanks for the informative section header informing us all of what your message would include—it was a nice prior warning.

First of all, most of the IPA templates do not link here, they link to help pages such as Wikipedia:IPA for English. So if you "came upon an article for a word you can't pronounce", this is usually not the page you would be taken to, you would be taken to a simple pronunciation guide.

Secondly, as numerous users have pointed out in the discussion above...even though you and a few other people might think "IPA is gibberish", if we used any pronunciation guide other than this then we would be going through this song and dance several times as often. IPA is the most common pronunciation guide in the world. If you think a particular article should include other pronunciations, leave a suggestion at that article's talk page and someone watching the article will have it. If you think all of Wikipedia should use a different pronunciation standard, take your suggestion to the relevant policy page to start a centralized discussion (and prepare to be told "no"). rʨanaɢ ^talk/_contribs 15:46, 5 March 2010 (UTC)

All those criticisms are based on a wrong statement: "English Wikipedia is only for native english speakers". For non native speakers those awkward spellings like "fŏŏt" or "bīd" used by the native speakers are just useless nonsense.

Sorry, the english language doesn't belong to you any longer, it belongs to the world. English wikipedia is the World wikipedia, not American or British.Japf (talk) 01:43, 17 March 2010 (UTC)

Font/Typeface for IPA

Maybe this is the wrong page for this comment but Please, Pretty Please can pages like this one use a decent font with clean and tidy display of IPA symbols? My preferred font for this is Lucida Sans Unicode, but maybe DejaVu Serif would be a better choice for Wikipedia. The font currently used generates an appalling mess for many of the things in the IPA chart (and almost justifies the crries of "incomprehensible" seen above).MichealT (talk) 12:42, 29 March 2010 (UTC)

The IPA template triggers several fonts depending on what your computer has installed. If you don't like the default that comes up on your computer, you can bring it up at {{IPA}} or else change your viewer preferences by setting up a personal CSS. (Explained at that link.) kwami (talk) 13:39, 29 March 2010 (UTC)

Thanks Kwami. I should have realised it was a my end problem, not a wikipedia one. MichealT (talk) 18:13, 29 March 2010 (UTC)

Need help

Can anyone translate Josef Zawinul's article and put the IPA translation into the introduction please? --Leahtwosaints (talk) 14:02, 12 June 2010 (UTC)

ːYou'd have to tell us how it's pronounced. "Tsa-vee-nool"? "Zah-win-əl"? — kwami (talk) 21:34, 12 June 2010 (UTC)

Consonant chart

I think the consonant chart needs reworking. First, it doesn't explain why there are two symbols in lots of places (unvoiced and voiced). Secondly, unless I am "blind," it is incomplete. ɕ and ɕ͈ and maybe more symbols are missing. This is really a problem for Wik users, since these symbols are used in pronunciation guides in Wik. Kdammers (talk) 01:34, 24 July 2010 (UTC)

The long standing good looking consonant chart was replaced by a horrendously ugly new one. The ideosyncratic styling is deviating from the general WP table styling. If it has to be a templete, then at least fill it with the previous layout. −Woodstone (talk) 04:04, 24 July 2010 (UTC)

RE Kdammers, the footnote on voiced--unvoiced pairs has a link to the article. A chart is not the ideal place to reproduce the IPA-handbook. On missing ɕ and ɕ͈ . ɕ Voiceless alveolo-palatal fricative (IPA-nr 182) is in the table Co-articulated consonants, row Fricatives. Isn't that where it is expected to be? And could you point where you'd expect ɕ͈ ? Name or IPA-number known? -DePiep (talk) 18:02, 26 July 2010 (UTC)

I suspect the person who touched the vowel chart ruined the consonant chart as well, does anyone know who did it and why they have to touch things which work well and have worked well for a very long time? The benevolent dictator (talk) 09:08, 31 July 2010 (UTC)

Vowel chart

Could someone explain why the vowel chart is now an image with it's own border and caption, and has moved from where it was to the right hand side of the page and from there to the right of the text which it used to sit below? I'm not wiki savvy enough to go back through the histories and find the original version of it but whoever touched it really screwed up. -The benevolent dictator (talk) 08:58, 31 July 2010 (GMT)

How the jump to "wiki-usage of IPA"?

Is there a solution/template/MOS/talk about introducing into this IPA-page a link, say: "about using IPA in Wikipedia"? A wiki-reader might be expecting that. -DePiep (talk) 20:57, 5 August 2010 (UTC)

Discovered: it's called self-reference (i.e. referencing to Wikipedia itself), see WP:SELF. Examples: Sandbox, Wikipedia,

Possible solution: use WP:HATNOTE. -DePiep (talk) 19:59, 6 August 2010 (UTC)

Asymmetry voicing vs aspiration (neutrality)

The following newly added section was removed quickly:

Neutrality

In IPA, aspiration and voicing are treated differently - voiced bilabial plosive is commonly written as [b] but not [p̬], while aspirated bilabial plosive must be written as [pʰ] when to distinguish with plain bilabial plosive. This cause that English speakers can simply use "[b]" and "/p/" (/p/ as a phoneme that can either be plain bilabial plosive or aspirated bilabial plosive) to refer their phonemes but a Mandarin Chinese speaker must use "[b] or [p]" and "[pʰ]" (Chinese people can never refer the phoneme that can either be plain bilabial plosive or voiced bilabial plosive in one character; nor can they refer aspirated bilabial plosive without diacritics). This phenomenon directly lead to the principal difference between Hanyu Pinyin and Wade–Giles.

The thought voiced in it has merit. The section is not written in perfect English and should perhaps not be a top level section, but the phenomenon is worthwhile mentioning. The opposition voiced–unvoiced is mostly marked by separate symbols. The opposition aspirated–unaspirated can only be indicated by a diacritic. That fits well with most European languages, where voicing is phonemic, and aspiration is not. There are languages (e.g. in India) where both aspects are equally phonemic. And as stated above, in Mandarin aspiration is phonemic, and voicing is not. The choices made in IPA are biased and applying IPA to Mandarin is less friendly by lack of plosives unmarked for voicing. −Woodstone (talk) 16:13, 21 August 2010 (UTC)

If something like this is to be included, someone needs to find sources to cite (rather than just writing their own analysis), and write it in proper encyclopedic tone rather than making an extended example out of one language. What you just wrote is shorter and more comprehensible than the text (copied above) that I removed. I am a linguist and a Mandarin speaker and I still couldn't tell what point that editor was trying to make with his little essay. rʨanaɢ (talk) 16:30, 21 August 2010 (UTC)

Yes. Sourcing is good. (And there ought to be a source that makes much this point.) The design of the IPA necessarily involved some arbitrary choices about how to represent the arbitrary differences among phonological systems of differing languages. It would be fine to mention something like this, as long as the statement can be sourced, as I suspect it can be. -- WeijiBaikeBianji (talk) 17:56, 21 August 2010 (UTC)

Agreed. The IPA is international in the European sense, and has grown only slightly since then. It's not just aspiration: various other phonatitons, ejectives, etc. are as fundamental as voicing, yet require diacritics. Also, although the IPA is supposed to cover all phonemic distinctions, it doesn't bother if they are from obscure non-european languages (voiceless implosives, nasal and voiced clicks, lateral fricatives), whereas european distinctions that can be written with diacritics get separate letters (ɕ ʑ ʍ ɧ ɶ).

I don't think we need any more citation, for those sources already cited in this article can be evidences of asymmetry voicing vs aspiration, and as "Where symbols appear in pairs, left—right represent the voiceless—voiced consonants" written in the article, we don't need any-more inline citation here. --虞海 (Yú Hǎi) (talk) 05:15, 22 August 2010 (UTC)

Of course you need sources for this; in general, I would question the way it's written as it doesn't seem to illustrate the shortcomings of IPA, but rather sounds, to put it bluntly, like a "complaint from China"... while I haven't dealt with IPA in years, I am fairly certain that there are also shortcomings for representing other languages; if such a section is included, a more diverse set of examples might be warranted. Maybe even calling that section "shortcomings," "deficiencies," or something like that. Choyoołʼįįhí:Seb az86556 ^{> haneʼ} 06:49, 22 August 2010 (UTC)

I do think we need a source, if only to avoid having everyone adding their pet peeve to the list.

Seb, it's not a matter of shortcomings for particular languages, but rather that the IPA reflects the languages it was originally designed for, and which the Latin alphabet was developed for. Thus voicing in plosives is elevated to a primary distinction, whereas aspiration, which is just as basic, is reduced to a diacritic, because Latin and major European languages have voicing in plosives but not aspiration to any great degree. If the ancient Greeks had designed it, all three would presumably have been accorded equal status, whereas the Chinese might have had separate letters for aspirates but a diacritic for voicing. The result is arbitrary, rather than being based on phonetic criteria. — kwami (talk) 07:05, 22 August 2010 (UTC)

Agreed, but this surely isn't confined to Chinese; that was my only point. Choyoołʼįįhí:Seb az86556 ^{> haneʼ} 07:12, 22 August 2010 (UTC)

Sure, Greek aspirates, Hindi murmured C's, Arabic emphatics, Amharic ejectives, etc. What surprises me is that there's a fictional consonant added just for Swedish, allegedly because there have been a lot of Swedes on the IPA board, but Swedish vowels cannot be adequately transcribed even with all the diacritics. — kwami (talk) 07:16, 22 August 2010 (UTC)

Can IPA or Extended IPA represent voices accurately?

VOT

e.g. In English /b/ is not fully voiced (and this might be a reason why English speakers often expect /b/ when a Mandarin Chinese speak out tenuis bilabial plosive; and the reason why most English teachers in China say a Chinese b and an English /b/ are merely "different on degree", no differences in nature). How to represent this in IPA? --虞海 (Yú Hǎi) (talk) 05:48, 22 August 2010 (UTC)

[p], as opposed to [pʰ]. rʨanaɢ (talk) 06:27, 22 August 2010 (UTC)

Transition consonants

In many language (espcially minority languages), we may find one or two consontant(s) that is "between" 2 IPA consonants (this time, in Mandarin Chinese we cannot find an exact or indisputable example, but pinyin character "h" might be a non-proper example), how 2 deal with this in IPA? --虞海 (Yú Hǎi) (talk) 05:48, 22 August 2010 (UTC)

IPA was never meant to perfectly represent every sound of every language; the only thing that can do that is an audio recording. IPA is a shorthand to help represent the important distinctions. People adapt it to fit whatever language they are using, and they use a broader or narrower transcription as is necessary. (For example, people transcribing American English often use [r] in place of [ɹ] because it's easy to write and the distinction between the two is irrelevant in Standard American English. Likewise, someone transcribing Navajo aspirated stops, which have a longer aspiration than English aspirated stops, might still use the diacritic ʰ because there's no need to contrast them directly; this diacritic is intended to represent the distinction between aspirated and non-aspirated within one language, not to directly record exact VOTs.) 你就是找碴儿。。。 rʨanaɢ (talk) 06:27, 22 August 2010 (UTC)

(after edit conflict, slightly overlapping)

This is not rare, even in major languages. In vowels it is even abundant. The common approach is to arbitrarily choose one of the candidates, giving preference to symbols that are plain latin letters or carry no diacritics. Actually that is exactly what Pinyin does. For the two phonemes "aspirated bilabial plosive" and "non aspirated bilabial plosive", two simple symbols are available /p/ and /b/ (disregardiing the [tʰ] and [dʰ] with diacritics). There would be no fundamental objection against making the same choice for IPA applied to Mandarin. Just as most English dialects have the phone [ɹ] and not [r], the latter /r/ is still almost universally chosen to represent the single rhotic consonant of English. −Woodstone (talk) 06:31, 22 August 2010 (UTC)

minor thingie:

"Some of the new symbols were ordinary Roman letters typeset "turned" (= upside-down) (e.g. ʎ ɥ ə ɔ ɹ ᴚ), which was easily done before mechanical typesetting machines came into use."

"before"? I don't get it... maybe "when mechanical typesetting machines came into use"?

Choyoołʼįįhí:Seb az86556 ^{> haneʼ} 07:01, 22 August 2010 (UTC)

Poorly worded, but 'before' is correct. When you lay out type manually, it's easy to flip a letter over 180°. You can't do that with a typewriter. — kwami (talk) 07:09, 22 August 2010 (UTC)

Yeah, I saw that now; I'll pipe it with a link to Hot metal typesetting -- I don't think everyone knows what that process entails... has me fooled there. Choyoołʼįįhí:Seb az86556 ^{> haneʼ} 07:13, 22 August 2010 (UTC)

a as in bake, ladle

Whenever I use a dictionary, it'll have a phonetic pronunciation, and then -- sometimes at the bottom of every other page, sometimes on the inside cover -- there'll be a little box that shows, you know, that ă is the vowel in star, far, car. I know significantly more about linguistics that your average person, but even I glaze over when I'm told that some symbol represents a "lateral dento-ventricular such-and-such." Just tell me it's the zh sound in azure. I'm not a linguist, after all, just a guy who wants to figure out how something is pronounced.

So, do we have this somewhere? I need this kind of pronunciation information on Wikipedia rarely enough that I'm not going to memorize the whole IPA system, but every once in a while I'll be reading an article like Great Vowel Shift and I wind up completely stumped. Hey, a left-pointing horseshoe followed by a weird colon made of triangles (ɔː) turned into a schwa followed by an upwards-pointing horseshoe with a little rainbow underneath it! It's like reading a chart of the history of Lucky Charms compiled by someone on meth. And so I dutifully trudge over to International Phonetic Alphabet, hoping that the page has improved since the last time I read it a couple of months ago. And you know what? I still can't for the life of me figure out what sound a left-pointing horseshoe followed by a weird colon is. On the vowel chart, the horseshoe is in the "open-mid" row and the "back" column, and the little colon (ː) doesn't put in an appearance as far as I can discern. So is it the vowel in "far"? In "man"? In "bet"?

When I read an IPA pronunciation somewhere on Wikipedia, I'd like to be able to go somewhere that allows me to convert the symbols into sounds I know within a few seconds. Just like dictionaries let me do! Do we have something that lets me do that? Calling it an "open-mid-back" vowel (plus weird colon) isn't cutting it, I'm sad to say.  Glenfarclas  (talk) 05:32, 4 September 2010 (UTC)

I guess the answer to my question is WP:IPA for English, a page I would never have come across by myself (I saw it mentioned in another comment here), that has an inscrutable title, and that is buried down in the see-alsos of this article without any explanatory remark. Can this be made more prominent somehow? I don't understand why it has to be so difficult and hair-pullingly frustrating to get to what is apparently the only place on the project that will explain to the average English-speaking Wikipedian, without frills, how to pronounce IPA symbols.  Glenfarclas  (talk) 05:47, 4 September 2010 (UTC)

I don't think you looked very hard, given that you found your answer only 4 minutes after typing up your long complaint. If you looked harder, you may have noticed that the top of IPA#Description has a clear link to IPA chart for English dialects, which would have told you that [ɔː] is the vowel found in "law, caught[24], all, halt, talk". The very top of the article also has links to WP:IPA (which would have told you the same stuff) and WP:IPA/Introduction. The phoneme articles you are complaining about (such as, for example, Alveolar lateral approximant) all have long tables which explain what words these sounds are found in, if you scroll down.

As for your argument about how dictionaries' pronunciation guides are "better", consider reading the numerous iterations of this dispute elsewhere on this page and on WT:IPA. Dictionary pronunciations are not standardized, so "zh" may mean something different to you than it does to me; IPA is an international standard. rʨanaɢ (talk) 12:04, 4 September 2010 (UTC)

re Rjanag. Your reply is missing the point Glenfarclas makes, thereby illustrating the need for an improvement as requested. Basically it describes that if you know where to search well enough into IPA, it's easy to find. The question was: why need for searching at all?

Minor thing: I calculate, the search took 05:47-05:32=15 mins, not 4. Add to that the unknown time seaching before typing the first entry. Quite enough to get hopeless. Another minor point: s/he does not state that the non-IPA guides are better, s/he says they are within easy reach.

Then, if IPA is an international standard, why should there be a WP:IPA for English or IPA chart for English dialects at all? Would the standard be different for Polish then? These two pages are titled wrong. It should be like: "Pronounciation of English [dialects] using IPA".

On [[IPA chart for English dialects]: it is an encyclopedic article, in the main space, and not the guide to pronounciation (the one which Glenfarclas found on the inside cover of a printed dictionary). Would you expect a pronounciation guide in the dictionary under "phonetics"? Such a guide should be, by definition, a Selfref(erence), i.e. about using Wikipedia itself. The top hatnote you mention, WP:IPA/Introduction, is a selfref indeed, but does not address the question (!). You need to click once more on the right link to arrive at Wikipedia:IPA for English#key (a section, mind you).

What Glenfarclas asks for, and what indeed is missing, is a single place where all pronounciation is explained (not: where IPA is explained) that is used in en.wikipedia. This place should be within one click of where IPA is used (even if when within a user signature like rʨanaɢ). Since we are electronic not printed, we could even expect a sample sound too with every symbol! A try is done here, using sound buttons, but it is not working gently. I think Glenfarclas has put us on a usefull track. -DePiep (talk) 13:47, 4 September 2010 (UTC)

This long message is nothing but a combination of nitpicking and incorrectness. Correction about timestamps: nitpicking. Title of WP:IPA for English: nitpicking (don't play stupid, you know what that page means). As for your claim that a "single place where all pronunciation is explained" is "missing", maybe you should take a look at WP:IPA, which I already linked once (and which is at the very top of the article). As for how IPA keys should be one click away, most IPA-related templates do just that (see, for example, {{IPA-fr}} and {{IPA-cmn}}). I don't really know what you're asking for, everything you're asking for already exists (in most cases, multiple versions of it exist to cater to your every need) and is already linked quite prominently.

In the amount of time you guys have spent complaining about nonexistent problems, you could have just learned basic phonetics and IPA already. It's really quite simple. rʨanaɢ (talk) 23:54, 4 September 2010 (UTC)

(out of timeline) nitpicking: yes, I already said "minor" there. Incorrectness: no. You with the as yet unspeakable IPA signature wrote: "IPA is an international standard" and "use IPA chart for English dialects". Together. Those IPA-smarties are indigestible. More constructive comments below. -DePiep (talk) 00:16, 5 September 2010 (UTC)

It seems like you are not a native English speaker, so I will explain this. "IPA chart for English dialects" means "chart containing the IPA symbols that are used to write English, rather than IPA symbols for sounds that don't exist in English". The IPA is standard, but not all of it is needed for every language. rʨanaɢ (talk) 00:41, 5 September 2010 (UTC)

If IPA is universal, there is no need for "IPA [...] for English". -DePiep (talk) 00:53, 5 September 2010 (UTC)

Read what I just wrote. rʨanaɢ (talk) 00:55, 5 September 2010 (UTC)

If the name "IPA chart for English dialects" means something else, then rename that page. There is no "IPA for English". -DePiep (talk) 01:07, 5 September 2010 (UTC)

The "IPA chart for English dialects" is an IPA chart for English dialects. The "IPA for English" is the subset of the IPA used for transcriptions of English on Wikipedia. We don't want to link someone to the entire IPA when they only need a fraction of it. I don't see anything wrong with either name. — kwami (talk) 01:10, 5 September 2010 (UTC)

re Kwami: The "IPA chart for English dialects" is an IPA chart for English dialects: no. It is (like) "IPA pronounciation by English dialects". That should be the name.

Then, the OP asked for something else. Please stop explaining IPA unsollicited. We know Wikipedia can explain IPA. -DePiep (talk) 01:39, 5 September 2010 (UTC)

No, the pronunciation of a given IPA character is the same no matter what dialect you speak; that's the whole point of the IPA. What differs from one dialect to another is the pronunciation of words; the IPA characters that are mapped to those pronunciations remain the same. rʨanaɢ (talk) 01:47, 5 September 2010 (UTC)

I can easily agree, if you change your first 'no' into 'yes'. I do understand IPA's aims. So my claim to by stays. It's not "an IPA symbol for an Australian", but (like) "an IPA symbol, as an Australian would pronounce it". Also does my statement stand: the OP asked for something else. -DePiep (talk) 01:57, 5 September 2010 (UTC)

You still don't get it. Everyone who pronounces [ɔ] pronounces [ɔ]; in other words, an open-mid back rounded vowel is an open-mid back rounded vowel. Speakers of some dialects may pronounce other sounds in place of [ɔ] (i.e., some people pronounce "caught" with an [ɔ] and some pronounce it with [ɑ]) but that is not because they pronounce [ɔ] differently, it's because they pronounce the digraph <au> differently in that word. If an American pronounces some word differently than an Australian , it's not because his [ɔ] is different than an Australian [ɔ], it's because he is saying something other than [ɔ]. [ɔ] is [ɔ] is [ɔ]. rʨanaɢ (talk) 02:25, 5 September 2010 (UTC)

Your analysis of what I was trying to say is correct. I know IPAophiles are very defensive about the IPA system, and my point is not to criticize it. It seems like a fine and thorough system. Indeed, any system used by another dictionary still requires me to look at a key to find out what, for instance, "ă" is supposed to sound like, so I don't mind it. But the key is pretty valuable information. For most people, it's the whole dang point, and all they need. That's why they put it on the inside cover of a dictionary or at the bottom of every other page, where it's very easy to find. Other pronunciation systems are not "better" (and I never said so); what's better is being able to find it and understand it without substantial effort and being a WP poweruser. The whole idea is that a pronunciation system should let you pronounce something, ideally without reading an hour's worth of linguistics or spending twenty minutes searching through random comments on the talk page until you stumble across one that happens to mention what you're looking for. I can assure Rjanag that I have come to the IPA page at least half a dozen times to try to decipher the pronunciation of something, and have wound up defeated and discouraged every time but the last -- and it's not like I'm some sort of Wikipedia novice, either.  Glenfarclas  (talk) 16:32, 4 September 2010 (UTC)

I think the solution has been well prepared by the creation of the template {{IPAc-en}}, which shows the applicable key examples by hovering over the symbols, and can be enhanced with a sound file if available. Coding {{IPAc-en|ˌ|æ|l|ə|ˈ|b|æ|m|ə}} leads to:

• Alabama: /ˌæləˈbæmə/

and {{IPAc-en|US|pron|audio=en-us-Alabama.ogg|ˌ|æ|l|ə|ˈ|b|æ|m|ə}} gives:

• Alabama: US: /ˌæləˈbæmə/ ^ⓘ

We should be more active in replacing the existing cases by this much more useful one.

−Woodstone (talk) 17:16, 4 September 2010 (UTC)

In your example here, that is a great cutting of the Gordian Knot. Now, please explain how to apply it. Please be specific, since a lot of us editors do not understand Wikbabble. For example, the poet Simić has this formidable-looking explanation: (IPA: [/ˈtʃ͡ɑːɻls ˈʂimitɕ͡/]). What are the steps to take to add a mouse-over alternative? Kdammers (talk) 00:05, 5 September 2010 (UTC)

I'd say don't blame good Woodstone for giving a solution. It even speaks! -DePiep (talk) 00:19, 5 September 2010 (UTC)

I'm not blaming her/him; I'm THANKING him/her. But I was also asking how to implement it in Wik articles.Kdammers (talk) 00:36, 5 September 2010 (UTC)

You are right, I'm sorry. I was too negative after reading an earlier comment above. So now I struck. Please go ahead. -DePiep (talk) 00:43, 5 September 2010 (UTC)

(ec) AFAIK, the template is only currently set up for English pronunciations. For other languages, someone will have to make a new template (probably using the same basic skeleton as IPAc-en, but replacing the recognized characters and the examples with ones customized for the given language). rʨanaɢ (talk) 00:45, 5 September 2010 (UTC)

I don't think that would be very useful, as it would require the reader to already have knowledge of that particular language. For example, would Kdammers really be helped by a popup that explains [tɕ͡] as the ć in Simić ? The transcription itself links to a key, just as you'd find at the bottom of the page in a dictionary. — kwami (talk) 01:04, 5 September 2010 (UTC)

Kwami: Yes, I would be helped by a pop-up that gives a common word using that sound. I'm not sure what you mean by " The transcription itself links to a key...." When I go to the IPA rendering of the name, which is blue-linked, and click on the symbol (which I cannot really see because it is so crammed together, and my eyes aren't the best), I get sent to Wikipedia:IPA, which is hardly the solution to my problem. To use that, I would have to copy the individual IPA letter and do a search in IPA. Copying individual letters in words -- especially blue-linked ones -- is difficult. Any-way, I actually did it, and went to IPA, but the search function simply led me to a discussion of the tie-bar. Come on, that is not at all like simply glancing down at the bottom of a page in a dictionary an dseeing that tɕ͡ is pronounced like K in keep or ch in such. Kdammers (talk) 03:01, 5 September 2010 (UTC)

I agree that the main IPA key is a barrier to reading the transcriptions. That's why we're developing shorter IPA keys for various languages. For the ones we have so far, see {{IPAkeys}}.

So, if you looked up a Serbian name with a [tɕ͡] in it, and you didn't know what that was, so you did a mouse over and got a popup with a common Chinese word that had that sound, would that do anything than frustrate? We could, of course, give English equivalents to sounds which do not exist in English, though I imagine there would be a huge fight on many of them: we have a hard enough time agreeing on that for English! Also, there would be a huge number of letter-diacritic combinations that wouldn't return anything, because we couldn't program IPA-all (Woodstone, below) for every eventuality.

But Serbian does have its own key. I'll change that article; no-one needs the English pronunciation of "Charles". (Tell me if it isn't now a lot more user friendly.) Also, by linking all the Serbian pronunciations to a central key, every once in a while someone will pass through and make sure they all agree with each other and with the Serbian phonology section, so that there's some consistency for the reader. — kwami (talk) 06:42, 5 September 2010 (UTC)

(outdent) It might be possible to make a template like {{IPAc-en}}, let's say "IPAc-all" that pops up the short examples as shown in WP:IPA and links to the individual line in that page, or to the full article describing the sound. −Woodstone (talk) 05:07, 5 September 2010 (UTC)

I think we'd need a different template for every language or language cluster, and it would be a huge amount of work. I've converted several tens of thousands of articles under the current system, and it's still a mess; I can't imagine doing several times that much work just so that readers don't have to glance over the Serbo-Croatian IPA chart. — kwami (talk) 06:52, 5 September 2010 (UTC)

bullets in chart

Under the vowel chart, there is a note: "Left and right of a bullet are unrounded · rounded vowels" I only see bullets outside of the chart, in the upper left-hand corner and in the cited text. I'm pretty sure this is not what's being referred to. I believe what is meant is to the left and right of a dash are unrounded and rounded vowels, respectively. If so, either the chart or the caption should be changed to reflect that.Kdammers (talk) 00:40, 5 September 2010 (UTC)

"bullets outside of the chart"" --no, that not correct! It should look like when you click "chart" for the pictural image (is what yo describe).

Could it be a browser thing? Here, on my WinXP, both Firefox and Safari show OK. -DePiep (talk) 00:47, 5 September 2010 (UTC)

(ec)These are bullets originally, but being wrapped in the {{IPA}} template and whatever other formatting commands are being used in that template seems to make them look like ndashes (at least on our respective browsers). Compare:

• {{·}} yields  ·
• {{IPA|{{·}}}} yields  ·

This is something that should be clarified, though. I would suggest leaving a message at Template talk:IPA vowel chart. rʨanaɢ (talk) 00:50, 5 September 2010 (UTC)

What I read is, that the bullets are outside of the chart. That's a placement thing, not a rendering thing. So I think CSS or browser (skin) first. Anyway, Talk expected at IPA vowel chart. -DePiep (talk) 01:00, 5 September 2010 (UTC)

Wait, is it about IPA vowel chart or template {{IPA vowel chart}}? -DePiep (talk) 01:02, 5 September 2010 (UTC)

I'm using FF. I see bullets at {{IPA vowel chart}}, but a hyphen in rʨanaɢ's example. — kwami (talk) 01:07, 5 September 2010 (UTC)

I'm using WinXP / Opera. In article view, I get the dashes; when I go to the source (i.e., when i click chart and wait for it to come up), it has the bullets. Kdammers (talk) 01:13, 5 September 2010 (UTC)

(ec, saving my edit first)Reducing confusion: if the bullets are OK in {{IPA vowel chart}}, then what's the problem? Kdammers sees them outside of the chart. That's different & serious. So far, the chart-template is not decided negative to me. -DePiep (talk) 01:14, 5 September 2010 (UTC)

By "outside the chart," I was referring to the bullets between the letters v, d, and e. The dashes occur when viewing Template:IPA vowel chart. Kdammers (talk) 01:18, 5 September 2010 (UTC)

The bullets outside the chart are irrelevant, DePiep. We're talking about the symbols between the vowels. rʨanaɢ (talk) 01:24, 5 September 2010 (UTC)

(edit conflict) I get it. So no placement problem. Now in my FF-on-WinXP, both yield-examples above are OK (true bullets) -DePiep (talk) 01:44, 5 September 2010 (UTC)

Instead of going to the trouble of worrying about font and browser differences, how about just clarifying the caption? Problem solved? rʨanaɢ (talk) 01:31, 5 September 2010 (UTC)

Is not what the OP pointed at, is it? The bullet went wrong in the chart itself, if I read well here. Then, the new footnote is leaving the (famous & well-known) IPA-text, and not changing the problem. And longer. Not my choice. -DePiep (talk) 01:50, 5 September 2010 (UTC)::Ah, and the footnote is less correct in case there is no a full pair. And is the rendering solved then? -DePiep (talk) 01:59, 5 September 2010 (UTC)

The problem was the the caption referred to bullets and on some browsers the bullets don't display as bullets. The caption no longer mentions bullets, so there is no longer a problem. rʨanaɢ (talk) 02:21, 5 September 2010 (UTC)

The OP read: I only see bullets outside of the chart, in the upper left-hand corner and in the cited text [=the footnote]. User:Kdammers, can you confirm the solution? -DePiep (talk) 15:04, 5 September 2010 (UTC)

I don't know what "OP" means, but I can condirm that the re-wording of the caption solves the problem admirably. Kdammers (talk) 09:55, 6 September 2010 (UTC)

OP=Original Poster=you, first post in this thread. I used it as "Original Post" (text), a bit off grammatically. But it's solved. -DePiep (talk) 10:40, 6 September 2010 (UTC)

In retrospect. Quite late, but it might be clarified by this: (a) the bullet in the footnote was not in a {{IPA}}, while the chart ones are. That might explain different fonts used (& outcomes visible) in the same template. (b) Then, to address kwami's note above (like "in my browser: the 2 examples here do not reproduce the stuff in the template"): the template is a table set to class="IPA". The examples are not. -DePiep (talk) 08:05, 7 September 2010 (UTC)

letter & symbol

IPA symbols are of two basic types, letters and diacritics. "Symbol", however, was used in the article for "letter", with resulting claims that sound do not have "symbols" when in fact they do. ("Diacritics may also be employed to create symbols for phonemes, thus reducing the need to create new letter shapes." International Phonetic Association, Handbook, p. 27) I've substituted "letter" throughout when that was the intended meaning. — kwami (talk) 05:52, 16 September 2010 (UTC)

This is good enough to be included in the article. May I suggest to write "IPA symbols are composed of two basic types, ..." (or: build from), which to me is more specific and more clear. Anyway, so a "symbol" is the composed unit, like k​͡​p and tʃʼ. -DePiep (talk) 10:20, 16 September 2010 (UTC)

"Symbol" can refer to a letter (e.g. [t]), to a diacritic alone (e.g. ʰ), or to a character composed of a letter and one or more diacritics (e.g. [tʰ]); it's not a particularly technical term.

@kwami: maybe "graph" would also be an acceptable alternative in some places, if we need to mix up the wording some? rʨanaɢ (talk) 12:02, 16 September 2010 (UTC)

-Yes, I read it like: "IPA symbols are composed of one or more elements of two basic types, ..." (which I understand is correct, but may not be easy reading).

-I prefer not to introduce another word like 'glyph' here. We should be able to describe it by IPA itself (also preventing OR).

-Another way could be: adding no more Alphabetic (say graphic) description, but introducing the Phonetic aspect of symbols etc. Like (going out of my depth clearly): 'a symbol is a phonetic thing'. Still, adding this to a good & correct symbol/letter/diacritic description. -DePiep (talk) 12:41, 16 September 2010 (UTC)

Rjanag, your recent unsigned, antisynchronic edit is disturbing. Am I to redo my reply? Are you making my already given reaction off the mark? Don't make me research your intention, you can write it out here, innit. -DePiep (talk) 21:51, 16 September 2010 (UTC)

I don't have any evil intention, I was clarifying my post to make it clearer. There's no need to create a fight just for fun. Jeez. rʨanaɢ (talk) 00:43, 17 September 2010 (UTC)

I didn't say 'evil'. Countertime edits are disturbing. Is what I wrote. It gives me the task to rethink my earlier answers &c. You could have signed and written it, which gives a logical sequence. -DePiep (talk) 00:57, 17 September 2010 (UTC)

I would also prefer not to introduce terms such as glyph, graph, or grapheme; the article is dense enough as it is! Everyone knows what a 'letter' is, and if they don't know what a 'diacritic' is, we need that term anyway. I'll try to fit in DePiep's wording to clarify that by 'symbol' we mean something more general than either. — kwami (talk) 22:45, 16 September 2010 (UTC)

OK. Eh, please don't expect "everyone knows" what a letter is (that is why people encyclopede. I for sure. Really, IPA is difficult for uninitiated. And, after the rites of passage, IPA is still mixing internally a lot of terms). Let IPA say it, and e.g. kwami write it. Not a big deal btw. -DePiep (talk) 22:59, 16 September 2010 (UTC)

That's why I linked "letter". I should have said 'everyone's familiar with the word letter'.

Okay, see how that reads. Because the footnote print is so small, we might want to split the footnotes into two sections, one in normal size for notes, and a seconding in smaller font for citations. — kwami (talk) 23:06, 16 September 2010 (UTC)

No, no small footprint. Into main text! Please: copypaste the first post here into the text, so in a 100% font (now don't smuggle font-size again into like "95%", kwami). Really, just a regular readable text should do (notes &c too). This said, I'll leave it here. I'm getting too pushing. But as I said earlier here: the content is worth adding. -DePiep (talk) 23:23, 16 September 2010 (UTC)

IPA-chart with audio

At last the sound samples are available in charts!

• IPA vowels chart with audio and
• IPA pulmonic consonants chart with audio

Unfortunately, an editor has proposed deletion. -DePiep (talk) 09:47, 17 October 2010 (UTC)

The pages are currently discussed in an AfD. (rephrased)-DePiep (talk) 11:44, 17 October 2010 (UTC)

First of all, no one proposed deleting the charts, they proposed deleting the silly transclusions you created in mainspace for no apparent reason.

Secondly, do not canvass. If you want to leave a note about a deletion debate at a relevant talkpage, you should do so by leaving a neutrally worded message, not something like "look at this great page, come save it from deletion". rʨanaɢ (talk) 11:19, 17 October 2010 (UTC)

You are right about the canvassing, my enthousiasm carried me away. Rephrased.

The nom clearly includes deletion of templates, all though not in their nominating first post. Finally, you using "silly" is useless. -DePiep (talk) 11:44, 17 October 2010 (UTC)

Clean up external links?

I find there are far too many links in this list. Should they perhaps be cleaned up a little? 129.242.182.190 (talk) 13:29, 4 November 2010 (UTC)

Gentium Plus

Gentium Plus has been released. Should be supported before Gentium by class=IPA. (I don't know where or how to modify that.) — kwami (talk) 20:04, 8 November 2010 (UTC)

Here. Needs an admin.—Emil J. 20:30, 8 November 2010 (UTC)

Thanks. Didn't know where to look. — kwami (talk) 21:48, 8 November 2010 (UTC)

Move NATO phonetic alphabet?

You are invited to join the discussion at Talk:NATO phonetic alphabet#Move?. — Joe Kress (talk) 08:20, 9 November 2010 (UTC) (Using {{Please see}})