Talk:International Phonetic Alphabet chart for English dialects/Archive 1

Why this article
This article is an (almost) direct copy, in IPA of the existng page SAMPA chart for English. It's not intended to rival International Phonetic Alphabet for English which aims to provide a scientific approach to English phonology. Rather it's intended as a quick guide for non-linguistic users as to how to interpret IPA coding they may come across in ordinary articles. rossb 22:16, 4 Mar 2005 (UTC)


 * Well, that's what the quick reference chart at the bottom of International Phonetic Alphabet for English was for. Do we really need both? --Angr 06:45, 5 Mar 2005 (UTC)


 * I think that even with the new quick reference chart, International Phonetic Alphabet for English is too big, too technical, and too off-putting for the casual user, who has seen an IPA transcription in a general article, has maybe never heard of IPA, and just wants to know what the pronunciation guide means. I came across the SAMPA chart for English article,and realised that the length and format of this is exactly what is needed for this class of user, who will be able to follow the link to  International Phonetic Alphabet for English if they're interested in delving more deeply. rossb 10:10, 5 Mar 2005 (UTC)


 * I personally like this article a lot because it's a lot easier to use than the other one.

Cameron Nedland 02:08, 19 January 2006 (UTC)


 * I don't see what the point of this, or the more advanced duplicate linked to above, really is. Why isn't it all kept in English phonology like other languages? It seems more like a useful tool for editors than an actual article.
 * Peter Isotalo 15:46, 2 December 2006 (UTC)
 * The point is to have a quick and easy reference tool without technical terms for the benefit of people encountering IPA transcriptions of English words in Wikipedia articles. It should probably be in Wikipedia: or Help: space though, rather than article space. —Angr 13:52, 3 December 2006 (UTC)

r
Why is the 'r' used (eg. in run, very)? I thought the whole point of having an *IPA* chart for English is to use the correct IPA characters for the phonemes found in English. The majority of English dialects use an upside-down 'r' (&#633;).


 * It all depends whether you're talking about a phonemic or a phonetic tranacription, and whether a broad trancription or a narrow transcription. I would expect the main use of this chart to be to help reader's understand the pronunciation guides given on various Wikipedia articles, which generally use a broad, phonemic (slashes rather than square brackets) transcription. If you're looking for a more detailed treatment, see International Phonetic Alphabet for English, which goes into much more detail. For the purpose of this chart, and fully within the rules of the International Phonetic Association, it's quite legitimate to use a generic character such as r, without attempting to go into the various allophones. rossb 18:00, 8 Apr 2005 (UTC)


 * Then we might as well use 'y' instead of 'j' for the palatal approximant. It would be about as accurate, and it might better "help reader's understand the pronunciation guides given on various Wikipedia articles". Besides, 'r' is not an allophone of '&#633;', nor is '&#633;' an allophone of 'r' within most dialects of English.  The only thing they have phonetically in common is that they are coronals (alveolars).  If this site is for broad transcription, then we don't need the 'dark l' (&#619;).

Bad problematic in AuE
I know that the length of the vowels in both 'cat' and 'bad' are different, as the vowel in 'bad' is usually long whilst the vowel in 'cat' is usually short, but when written in broad transcription they are treated as the same vowel in AuE. So how is 'bad' problematic in AuE?

The pronunciation of /t/ in kitten is more like a glotal stop than a /t/, perhaps find a better example. – AxSkov ( T ) 12:16, 3 May 2005 (UTC)


 * Bad is problematic because for many Australians it has a phonemically long : it fails to rhyme with lad, which has phonemically short . For those speakers, it's not just a question of vowel lengthening before voiced consonants, since some 's are long before voiced consonants, while others are short in the same environment. So lad is better because it's definitely phonemically short and so a better correspondant to phonemically short RP (GenAm doesn't have phonemic length anyway). As for kitten, we use slashes on this page rather than square brackets, which means we're using a phonemic transcription, not an allophonic one. But we can change it to something else if you like. How about bidden ? --Angr/comhrá 14:08, 3 May 2005 (UTC)


 * Ok, use 'lad' instead. Yes, 'bidden' sounds like a far better word to use. The word 'ridden' is a better word to use rather than either bidden or kitten. – AxSkov ( T ) 09:58, 5 May 2005 (UTC)

Glottal stop
Kitten in American English is most definitely pronounced with a glottal stop. See The Glottal Stop in English. For non-British speakers who may not be familiar with the Cockney accent (myself included), this is a helpful example. - Barfooz 07:05, 5 May 2005 (UTC)


 * Kitten is pronounced with a glottal stop by some Australian speakers as well, and you'll probably find that some Candians and New Zealanders pronounce it that way too. I'm an Australian and I usually pronounce kitten with a glotal stop. – AxSkov ( T ) 09:53, 5 May 2005 (UTC)


 * Many speakers of American English use a glottal stop, but it's allophonic (not phonemic). Mo-Al 20:27, 7 August 2006 (UTC)

US vowels
Changed a few details on US vowels per Western US pronunciation (Ladefodeg at UCLA, etc.). "aj" has two phonemic realizations (rider vs. writer: the consonants are the same!) Bird has a syllabic ar, not a rhotic vowel. See and sue are diphthongs, though they're a bit subtle to hear (consider pill vs peel, where the latter sounds almost like a syllable and a half). Father has a long vowel (as we can see that the th isn't ambisyllabic like it is in feather), and the nucleus of bird is as long as that of bard, so it's best marked long too. [&#652;] is found in Vietnamese, but not standard US English, so I replaced it with [&#592;]. (Yes, [&#652;] is the standard transliteration, but using it for this sound is unique to English, so the IPA's no longer the International Phonetic Alphabet if we allow it.) Law is only [&#596;] in northeastern dialects, so I listed the "standard" pronunciation as well. Here is probably closer to [&#618;&#633;] than to [i&#633;]. The difference isn't great, but then we're listing all the other vowels before ar as "lax".

Could make a comment that [&#650;&#633;] is unstable for many speakers.

Initial ar isn't [&#633;]: It's labialized. Which is what enables us to pronounce rural: first ar's labialized, second is not:. True for all ars: raw. I didn't change this, but maybe a comment should be made.

Also should do something about secondary stress. There is no such thing, at least in most of the US. The primary/secondary/tertiary/quaternary stress distinction is due the the reduction or non-reduction of vowels, plus intonational stress. In battleship, the last vowel is not reduced, but not stressed either. (In traditional transcription, it would be given tertiary stress, not secondary.) In arachnophobia, the pho only appears to be more stressed when the words spoken in isolation. Say Arachnophobia's playing at the Bijou and the effect disappears. (That is, it's not really part of the word at all.) --kwami 23:38, 2005 May 16 (UTC)


 * This page is intended as a guide to IPA symbols that Wikipedia users are likely to encounter in articles on nonlinguistic subjects. The symbols used should be those that are most commonly used, they should be broadly phonemic, and they should be simple. This page is not the place to discuss fine details of phonetic realization. Likewise, secondary stress marks should be included because pronunciation guides are likely to include them, since article names will usually be pronounced in isolation. --Angr/comhrá 23:55, 16 May 2005 (UTC)

Pronunciation of 'with'
I agree with Woodstone's removal of with from the set of example words, but not entirely with his edit summary. The pronunciation of with is extremely variable across English accents: some people say in all environments, some people say  in all environments, and some people say  in some environments and  in others. At any rate, it's far too unstable in its pronunciation to be useful as a headword for either or. --Angr/undefined 20:57, 25 Jun 2005 (UTC)


 * If this helps at all, in NW Kansas says it [wI&#952;]Cameron Nedland 04:57, 7 May 2006 (UTC)


 * My dad on the other hand, who is from Northern Wisconsin, says [wɪð]Cameron Nedland 16:35, 4 November 2006 (UTC)

"tour" is another bad example
American English pronounces this as, not , so it's a bad example to use to illustrate. I can't think of any AmE word that uses ...I don't think you can list as the regular rendering of  either because "poor" is not pronounced with .... DopefishJustin (&#65381;&#8704;&#65381;) 20:29, July 27, 2005 (UTC)
 * There's no difference between and  in American English. The vowel is about halfway between  and  and the distinction between them is lost before . The transcription  is just a convention, but a long-established convention. --Angr/undefined 20:41, 27 July 2005 (UTC)

Merging from English phonemes
English phonemes tries to be a bad version of this article. The suggestion is more that we make it a redirect to here, rather than anything else. If no-one objects in, say, a week, I'll do the magic. —Felix the Cassowary (User talk:Cassowary) 03:47, 6 September 2005 (UTC)
 * Support merge. I have just had a look at English phonemes and I agree that the article is a poor imatation of this article, and therefore should be redirected to this article. – AxSkov ( ☏ ) 06:40, 6 September 2005 (UTC)
 * Support the merger. --Angr/undefined 06:12, 6 September 2005 (UTC)


 * I've done the merge. —Felix the Cassowary (User talk:Cassowary) 06:32, 10 September 2005 (UTC)

Bad-lad split
Some one removed this from the article, but I think it should stay, because Australian English does have a different phoneme in bad than in lad. 64.200.124.189 03:20, 19 September 2005 (UTC)


 * I don't think either this or the cot-caught-merging variety of American English should be included on this page. This page is not the place to include all possible varieties of spoken English; this page is an introduction to IPA symbols as used for English for the benefit of nonlinguists using Wikipedia. This page links to General American and to Australian English phonology, which mention the cot-caught merger and the bad-lad split (and have links to those phenomena). I think this page needs to be kept as simple as possible and not confuse the reader by listing all possible variations of pronunciation. But we can link directly to to cot-caught merger and bad-lad split from here. I'll do that now; tell me what you think. --Angr/undefined 08:37, 19 September 2005 (UTC)


 * I’m quite happy to leave the bad-lad split off here. Could I also suggest we remove, and  which I think also only get in the way of a good understanding. —Felix the Cassowary (User talk:Cassowary) 10:54, 19 September 2005 (UTC)
 * No objections from me there! --Angr/undefined 14:39, 19 September 2005 (UTC)
 * I don't know. I think it's useful having at least and  mentioned. – AxSkov ( ☏ ) 05:08, 20 September 2005 (UTC)
 * Why? Remember the main point of this chart is to help nonlinguists decipher the symbols they encounter as pronunciation guides in Wikipedia articles. Since those pronunciation guides invariably use phonemic notation, and  won't be there. --Angr/undefined 06:13, 20 September 2005 (UTC)
 * That's a fair point, so I won't object to their removal. – AxSkov ( ☏ ) 06:51, 20 September 2005 (UTC)

Canadian English
I'm removing the Canadian English column because as yet the article Canadian English has only a single paragraph on phonological phenomena in the accent, but no in-depth discussion of phonemes. Also, while Received Pronunciation, General American, and General Australian (as described at Australian English phonology) are relatively monolithic accents (not completely, of course, but to a certain extent), there are many varieties of Canadian English (as indeed there are of British English and American English, which is why those labels are not used here). Finally, remember that this article is intended as a guide to the use of phonetic characters encountered in the pronunciation guides of Wikipedia articles. Canadian English is close enough to GenAm that no characters encountered in Canadian-English pronunciation Guides will be confusing to some consulting the GenAm column of this page. --Angr (t·c) 15:02, 25 November 2005 (UTC)


 * I don't think that's fair for the Canadians.

Cameron Nedland 02:08, 8 December 2005 (UTC)


 * Especially for me, since I'm Canadian and I put those vowels in by analysing my own voice. Denelson83 19:47, 28 November 2005 (UTC)


 * If you based those transcriptions on your own voice, it's original research and can't be included anyway. If you want to do a phonological analysis of Canadian English, find out what the published sources have to say, and write it up at Canadian English. --Angr (t·c) 22:28, 28 November 2005 (UTC)


 * C'mon, Angr! I want to know what Canadians sound like.

Cameron Nedland 02:08, 19 January 2006 (UTC)

The "a" and "backwords c"
Those two sound the same to me, and I consider myself a "General American" speaker. Cameron Nedland 02:10, 8 December 2005 (UTC)
 * Then your accent includes the cot-caught merger. General American is usually defined broadly enough to include both varieties with and without the merger. --Angr (t·c) 23:56, 26 November 2005 (UTC)
 * Gotcha.Cameron Nedland 04:56, 10 August 2006 (UTC)
 * Looking at the above exchange, I'm curious how I was able to answer your comment almost two weeks before you made it. User:Angr 05:43, 10 August 2006 (UTC)
 * I probably did it before signing in and then later came back and signed it.Cameron Nedland 20:10, 10 August 2006 (UTC)

Rhotic diphthongs
Is there any particular reason the "or" diphthong was left out? It should be there (see Horse-hoarse_merger). I added it. Also, to be fair to American readers, I added "ar" (as in car) to the diphthong section. Another change was to the second letter in the rhotic diphthongs - using, a consonant, in a diphthong ( instead of ) makes about as much sense as using instead of. Of course, if I'm wrong, feel free to correct me, and give your reasoning. --EthanL 15:19, 3 December 2005 (UTC)
 * More technical matters like these are better handled elsewhere, like International Phonetic Alphabet for English or even R-colored vowel. This page is really supposed to be a quick and easy guide to IPA characters as used in Wikipedia articles for the benefit of those who aren't already familiar with them. The r-colored vowels are almost always transcribed with rather than . Also,  and  were left off because their RP and AuE equivalents aren't diphthongs, and it's confusing to have  and  in a list of diphthongs. Americans who encounter the transcriptions  and  can figure out what they mean by putting the parts together ( +  and  + . --Angr (t·c) 16:13, 3 December 2005 (UTC)

Other Dialects of English
Could somebody maybe add Cornish, Southern US, New Zealander, Irish, etc...? Cameron Nedland 20:12, 24 December 2005 (UTC)
 * This page is primarily to help users who are not familiar with the IPA read the phonetic symbols they encounter in Wikipedia articles. It is not the place to describe every dialect of English. --Angr (t·c) 20:29, 24 December 2005 (UTC)
 * This actually sounds like a good idea though. Where would be the place for a sound to sound comparison like that? would that not be an interesting and useful resource? --86.135.217.213 19:39, 10 January 2006 (UTC)
 * I'd say the place for a sound-to-sound comparison of the major dialects of English would be a doctoral thesis in sociolinguistics, not an encyclopedia. --Angr ( tɔk ) 20:04, 10 January 2006 (UTC)
 * I meant one on the scale of what's presented here. Just adding a few extra columns. If not here, maybe a new page?--86.135.217.213 20:30, 10 January 2006 (UTC)

Are they the same? Which one should I use? &mdash; Yaohua2000 10:19, 11 January 2006 (UTC)


 * For use in English, there is no difference. I think the first is the recommended one. (In some languages that have a phonemic contrast between the affricate and the sequence, the first one is used for the sequence, the second for the affricate. But that doesn't happen in English. In any case, the IPA recommendation is the tie-bar format, because there's only so many ligatures, and they're not necessarily obvious.) —Felix the Cassowary 10:32, 11 January 2006 (UTC)

Syllabic consonants
Shouldn't there be some information in this chart on how to make a syllabic consonant symbol? I for one have no idea how to do so, nor can I find out how anywhere on Wikipedia. &mdash; 172.162.28.72 02:05, 19 January 2006 (UTC)

Sounds like a good idea. Cameron Nedland 02:08, 19 January 2006 (UTC)


 * The symbol for a syllabic consonant is given in the last table in this article, called "IPA: Other symbols used in transcription of English pronunciation". Probably the simplest way to to make it is to type &amp;#809; after the consonant you want to mark as syllabic. Keep in mind the only consonants that are usually syllabic in English are /l/ and /n/; /m/ can be occasionally, but not as often as some people think. (Despite the spelling, the /m/ in rhythm isn't usually syllabic, as there is a brief /ə/ between the release of the /ð/ and the beginning of the /m/.) --Angr ( tɔk ) 06:05, 19 January 2006 (UTC)

RP vowels
To the anon who reverted my changes with the word "rubbish": they certainly are used in mainstream dictionaries. Four of them are used in the New Edition transcriptions in the OED, and similarly in the New Shorter Oxford - about as mainstream as you can get. See the OED pronunciation key. The use of instead of  is found in Collins, for example, and is probably more "standard" than .--JHJ 11:22, 21 February 2006 (UTC)
 * Maybe, but your not going to use both are you? For example: the transcription in an article, say bed, would be, using one, not both. You would not write both and , this is just confusing for the reader, it does not help them with pronunciation. For this reason keep it simple.  I cannot see JHJ's point with his link, it is not helpful for his case. Mark 11:41, 21 February 2006 (UTC)
 * Yes you are going to use both. The reader may be trying to interpret IPA, in which case it helps to have both possibilites listed. And when writing it, it helps to know your options. Having two options listed is hardly confusing. &mdash;Pengo 11:59, 21 February 2006 (UTC)
 * No you are not, one is enough. This is why there are other articles for this sort of thing. This is not the place for these differences within RP, if any. And of coarse it will be confusing for the reader. Mark 12:08, 21 February 2006 (UTC)
 * The OED link shows the other four symbols (not ) that you reverted - look at the transcriptions they use for the relevant vowels. As to your other point, of course I wouldn't use both transcriptions of the same vowel in the same article.  But the same applies to the double symbols for two GenAm vowels, which I note you didn't remove, and the fact remains that people are likely to find the alternative transcriptions in other sources and use them here - for example English phonology uses  in several places for the bet vowel.  There is a case for having some sort of clear standard, but if so I'm not convinced that we were using the right one (it matches neither Collins nor Oxford), there should be a conversion page somewhere for the different symbols found in other sources, and there should be a standard version for GenAm too.--JHJ 12:05, 21 February 2006 (UTC)
 * You're right that GenAm should only use one, but I didn't know which one should be used, so I left it alone. For RP, I suggest to you to pick one and only one to use. This article is not the place for alternative transcriptions. The English phonology article you mentioned is dominated by GenAm using /e/ in place of  and /o/ in place of . The /e/ you see is AuE. Mark 12:26, 21 February 2006 (UTC)

Repeating a comment made at Mark's user talk page, one way forward might be to create a page with a table with the alternative transcriptions, with a clear link from the chart page there.--JHJ 12:42, 21 February 2006 (UTC)


 * How about adding a bunch of footnotes to the table (i.e. list only one glyph, but have the other as a footnote). Pengo 13:13, 21 February 2006 (UTC)


 * OK, I've tried that out for three of them. If people are happy with it similar footnotes can be added for Oxford's alternatives to  and  and for the two American vowels with alternative forms.--JHJ 17:43, 21 February 2006 (UTC)


 * Thank you for doing that. I think it does actually look less messy, as well as being more informative. And hopefully it will put an end to the revert wars. &mdash;Pengo 00:47, 22 February 2006 (UTC)


 * I agree it looks much better now, the tables have been kept simple with also some additional notes, well done for this solution. I also agree that this solution should be done for the two American vowels as well. Mark 10:06, 22 February 2006 (UTC)

x is for Chanukah?
I noticed that in the consonant chart, by the letter x, the entire word "Chanukah" is bolded. I'm guessing that only a certain consonant sound is supposed to be bolded here. I would change it myself, except I don't know which letters exactly to bold. Dansiman 08:04, 14 March 2006 (UTC)
 * Fixed. Angr/ talk 10:37, 14 March 2006 (UTC)

Reorganisation into alphabetical order?
As this is a reference chart for those uninitiated in phonetics, I would consider re-organising it into alphabetical order for ease of use, like many dictionaries (rather than the current form in traditional IPA order of articulation, etc: the full chart is more useful for those who are interested), keeping consonants, vowels and diphthongs seperate but placing together all the similar sounds (i.e. all the "a"-like sounds, etc): this would make it much easier and quicker for users to find the transcription they are looking for.

—Djbb2 11:01, 27 April 2006 (UTC)


 * I'm not sure that it's necessary: dictionaries are organised alphabetically because they're very long and need to be indexed somehow, while this chart easily fits inside one or two screens (depending on your equipement). As a point of curiosity, how would [ʒ] or [j] be ordered (being two examples that are a bit problematic)? &mdash; Saxifrage &#9998; 12:02, 27 April 2006 (UTC)


 * Alphabetic ordering doesn't make much sense, since many of the symbols do not occur in the alphabet. Any assignment in the sequence would be arbitrary. However, I think the table should be reordered slightly to follow the rows in the IPA chart. For vowels perhaps better by column. &minus;Woodstone 17:05, 27 April 2006 (UTC)


 * Sorry, I wasn't clear. I meant some dictionaries' pronunciation keys, which organise the symbols "quasi-alphabetically" according to their similarity to Latin letters: [ɑ], [æ], [aɪ], [aʊ], [b], [d], [dʒ], [ɛ], [ɛə], etc. (based on the Hutchinson Encyclopedic Dictionary's key). To use Saxifrage's examples, [ʒ] would come after [z], and [j] would come after all the "i"-like symbols and [k]. For signs that bear no resemblence to Latin letters, the English orthography could be used, e.g. [s] would come before [ʃ], [t] before [θ]. I agree that there would be some arbitrariness, but most IPA symbols look similar to one Latin letter or another.


 * I think it's confusing for people who want to find a quick key to the pronunciation of a word to see "p", "b" and "t" at the top of the list; ordering the symbols quasi-alphabetically would make it quicker for them to find the one they are looking for. — Djbb2 21:48, 28 April 2006 (UTC)


 * Okay, that does address the ordering issue. Still, I don't see that reordering it is necessary since it's not difficult to search the list. The people that are going to be unfamiliar with the conventional IPA ordering are also unlikely to be using this very frequently, and if I learned nothing else from my programming classes, it's that you don't optimise anything to save time in infrequent cases. &mdash; Saxifrage &#9998; 08:56, 29 April 2006 (UTC)

If we're going to resort it based on such principles, I'd suggest going for the most common English spelling of the sound as the basis, so something like: æ (short a) æi (long a/ai) aː (short ar) eː (long ar/air) b d e (short e) iː (long e/ea) ɜː (short er) ɪə (long er/ear) etc. This would make it easier on someone unfamiliar to find the right IPA character, means that dialectal differences in sounds won't change the ordering, and means we don't have to decide if ʌ is a variant of v or a novel character that belongs with the U's. —Felix the Cassowary 14:11, 29 April 2006 (UTC)
 * Ack, no! If we did that, in the list you started above, we would have ...b k d..., which make finding things much more difficult, not easier. Angr (talk • contribs) 14:15, 29 April 2006 (UTC)


 * Not only that, but which pronunciation of spellings would you choose, British or American? Wikipedia is not allowed to favour either. &mdash; Saxifrage &#9998; 23:40, 29 April 2006 (UTC)


 * Good point Angr; I intended it to be implemented with common sense and only really that way according to vowels, so that /k/ should indeed go between dʒ and l (and we'd put the English orthography in the first column so that you're looking at "ch" and seeing it's tʃ, not looking at tʃ and seeing it's between b and d).
 * Saxifrage, I'm not sure how it is any more biased than the original suggestion. Differences in spelling between British and American are incredibly minor. -our vs -or isn't going to change the location of ə/ɹ, and s/z falls under the category of commonsense, so even if analyse vs analyze is enough it won't make a difference anyway. Differences in pronunciation (and so letter shape) are much greater, so əʊ vs oʊ will have different locations/a bias.
 * But I'm much happier to leave it in the current order anyway. This just seemed a better idea if it were to be reordered.
 * —Felix the Cassowary 00:39, 1 May 2006 (UTC)
 * I'm much happier to leave it as-is too. I think these proposals are too thorny and ad-hoc when we have an ordering standard already available. &mdash; Saxifrage &#9998; 00:52, 1 May 2006 (UTC)

[ts] affricate
Is the "ce" in "sentence" the [ts] affricate?Cameron Nedland 05:00, 7 May 2006 (UTC)
 * Certainly not at the phonemic level, since English doesn't have an affricate /ts/ phoneme. At the phonetic level I suppose a case could be made for it. Angr (talk • contribs) 08:22, 7 May 2006 (UTC)
 * Ok.Cameron Nedland 21:13, 15 May 2006 (UTC)
 * In your dialect it may well be [ts]; in many others, including mine, it's simply [s]. Both are valid pronunciations. 84.70.184.95 20:20, 1 July 2006 (UTC)
 * I'm glad I'm not the only one.Cameron Nedland 04:05, 5 August 2006 (UTC)
 * In mine (Australian English), at the phonetic level it's the affricate [ts], but at the phonemic level it's simply an /s/. So for example, the word 'chance' is phonemically and phonetically, hense [ts] is an allophone of /s/ when it comes after an /n/. In my dialect it is impossible for the sound [s] to purely follow the sound /n/, so 'chance' is always  never . If your dialect has this sound then you can hear a subtle [t] sound between /n/ and [s], when after making the sound /n/, the tongue leaves the palate to make the sound /s/. Marco 03:02, 24 December 2006 (UTC)

California English
Shouldn't California English be represented in this too? It's one of the largest varieties in the world, far bigger than Australian English. - Gilgamesh 17:11, 1 June 2006 (UTC)


 * What phonemes are in California English that aren't in this chart? &mdash; Saxifrage ✎ 17:39, 1 June 2006 (UTC)
 * California English can be subsumed under General American for purposes of this chart. Angr (talk) 06:02, 2 June 2006 (UTC)
 * Okay. - Gilgamesh 07:18, 2 June 2006 (UTC)

Linguistically detailed
User Angr's more easy to understand edit of the article may be true; but the earlier version to which Angr understandably had reverted was less clearly indicating that IPA stands for International Phonetic Alphabet and the link to it didn't stand out so well. There is an immanent reason: see Wikipedia talk:Manual of Style (pronunciation). Without the change I made in 'IPA chart for English', linguists would not be very happy if the suggested link from under any 'IPA' would redirect to this 'IPA chart for English'. As it stands now, I hope it will be acceptable for scientists and allows a highly needed simple solution for the laymen -- SomeHuman 2006-06-25 21:30 (UTC)


 * I'm not seeing how "liguistically detailed" adds anything (except confusion, since "linguistic detail" is not a meaningful term in linguistics and is going to be near-meaningless to someone who isn't familiar with linguistics enough to know how the word detail is modifying the meaning of linguistics...). As for your concerns at Wikipedia talk:Manual of Style (pronunciation), I answered that there and I summarise here: this edit does not improve the article, rather improves some extra-encyclopedic concern. Improvements to the article are the only reason to edit the article. &mdash; Saxifrage ✎ 05:42, 26 June 2006 (UTC)
 * Most people realize that 'linguistic' roughly means 'related to (the study of) languages', so the details have something to do with languages, right? The end of the line states that the current page on the other hand, relates to sounds in English. You know, that's what we readers read here. If we do not feel up to a study of languages, we better start reading the current page, then. Of course, if one is interested in a thorough understanding of IPA... – Do you honestly think people can not figure this out? How do you figure the Suggestion for practical IPA usage in Wikipedia to be an extra-encyclopedic concern? "Improvements to the article are the only reason to edit the article", so improvements to the encyclopedia are no reason to edit the article? -- SomeHuman 2006-06-26 12:20 (UTC)
 * A dictionary does not bootstrap its pronunciation guide by co-opting the "pronunciation" entry, and neither does an encyclopedia. If we need a separate project page about pronunciation, I invite you to propose or write it.
 * As for the substance of the edit, what has been added that the "concise..." phrase doesn't already say? &mdash; Saxifrage ✎ 19:44, 26 June 2006 (UTC)
 * The earlier phrase allows assuming, even suggests, that IPA is something for English (of which a concise version is underneath). The following line becomes then obscure: one has to read the first line again and realize that one's first assumption was false. My phrase clearly states that IPA is far more general, the page underneath is a version for English, the next line smoothly continues to point out there is a more comprehensive one for English as well. My phrase allows realizing that the link to the general IPA article is rather for linguists. And it shows the (IPA) abbreviation from which a user might arrive there directly besides the meaning. It makes it more suitable for the suggestion put in an already existing project. All that is gained by simply rephrasing. And if you do not see the difference, how valid can your objection to it be? -- SomeHuman 2006-06-28 00:06 (UTC)
 * It's not that I see the difference, it's that I do not see the advantage. And no, a suggestion in one project of Wikipedia does not warrant mucking with articles. Of course it's for linguists: the IPA is not useful to anyone else, concise or otherwise. The associated suggestion of pairing spell-outs with IPA in articles that need a pronunciation guide is sensible and doesn't require mangling Wikipedia's linguistics articles to their detriment. &mdash; Saxifrage ✎ 08:45, 28 June 2006 (UTC)


 * Acutally, the IPA is also useful for stylish rabbits and poor pig immitators >:ʙ :ʘ —Felix the Cassowary 11:40, 28 June 2006 (UTC)
 * Linguists rule the World, eh? There may be very few people besides them to write the IPA symbols but most passive users are not linguists at all: just people who are curious to find out how an unfamiliar word or name is pronounced. Perhaps an average American might not be aware of it but a linguist, even from the US, should realize that dictionaries in England – often preferred by non-native speakers as well – and dictionaries for other languagues very often have the IPA behind the entries. Thus a lot of users on earth are IPA-wiser than the 4% of the world's population in the USA. With current attention for the Internet, there are probably already more speakers of another language using the English language Wikipedia than American users. Saxifrage does not even attempt to disprove my 1st argument: avoid suggesting IPA is for English language only (Is a linguist interested in pronunciation no longer aware of grammar?), or my 3th argument: show the abbreviation immediately behind its written-out meaning. The latter might get in the other text just as well, but Angr does not bother to improve, just keeps reverting as quickly as possible to a version for which no arguments at all have been provided to demonstrate a supreme value. My arguments stand regardless whether the 'IPA chart for English' (the least linguistical about IPA) should be the destination of a multitude of links or not. -- SomeHuman 2006-06-28 16:00 (UTC)
 * My reverts are based on the fact that your version (1) is syntactically very poor English and (2) doesn't add any useful information to the original version. If you have a problem with the fact that the IPA2 template links here (or used to, it doesn't anymore), the place to take that up is Template talk:IPA2, not here. User:Angr 16:11, 28 June 2006 (UTC)
 * The added information would not have been useful to you. If the page is for linguists only, it should not even exist on Wikipedia. Since you criticized the syntax, why not improve it then – in such way everyone realizes that it's a concise IPA version for English while not allowing anyone to assume that a comprehensive IPA version for English would constitute the entire IPA. -- SomeHuman 2006-06-28 17:03 (UTC)

My question is why would a comprehensive International Phonetic Alphabet version for English constitute the entire IPA? I mean, maybe some few people think English is the only language used throughout the world, but I would hope the vast majority are not of that delusion. Also, the fact that it says "for English" should indicate that there could be an IPA for French or for Pitjantjatjara. If the IPA was only for English, why would anyone specify that in this article (where it would be redundant) instead of on the IPA page (which is linked to very prominantly from this page). Users must be expected to follow the links that are provided; no page can provide every single bit of information about something, and nor should they try to. I quite like the present/original version. It is clear and concise, and I know exactly what it means. All the other posibilities have left me feeling much less happy. My only other suggestion is changing "for" to "as it is used to describe", which is longer but perhaps less ambiguous. —Felix the Cassowary 13:25, 29 June 2006 (UTC)
 * It's possibly ambiguous, but I don't think it's actually ambiguous. In this case and most cases, the more likely interpretation of "X for Y" is that Y modifies X: consider "food for dogs", "Internet for Dummies", "a gift for you", and the like. None of these are interpretable as suggesting that food is specific to dogs, Internet specific to "dummies", or gifts specific to "you". Thus, if the article really meant that IPA was for English only, it would be titled in a different way than a structure that is discoursively unambiguous (even while being syntactically ambiguous). My lack of response to the argument when SomeHuman presented it, then, was because I don't think it's substantive.
 * To sum, I believe the changes are unnecessary on grounds of clarification and no other convincing demonstration of their utility has been made to justify the increased awkwardness of the phrasing. On an unrelated note, I see no problem with trivially adding the acronym IPA after International Phonetic Alphabet. &mdash; Saxifrage ✎ 17:18, 29 June 2006 (UTC)


 * Mentioning "an IPA for English" is an oxymoron. It is "the IPA as applied to English". That concept should be very clearly conveyed to the reader. Furthermore, according to IPA rules, there is more than one correct way of applying the IPA to English. So indeed adding the words "concise" and "most common" is of value. &minus;Woodstone 17:18, 29 June 2006 (UTC)


 * The IPA, like any 'alphabet', is a list of symbols. Assume a user not knowing what 'IPA' is about – the page is titled 'IPA chart for English', remember – clicks on some link and sees this page. Of course he knows what an alphabet is: a list of symbols. So the user understands "This is ... the List of Symbols for English sounds". The predicate 'International' then must mean: fit to represent American, Australian, Scottish,... English, so indeed the alphabet needs to be a phonetical one to represent the different sounds in the dialects. This is not a stupid user. Only one who already knows enough about the specific phonetic system named IPA will not be misguided. Saxifrage makes a crippled comparison because everyone knows the internet, food, etc. His preferred syntax is only syntactically valid (unambiguous) when that condition is guaranteed. The next line continues to specify English each time, as if the editor was carefully preventing a reader to think IPA might be designed for any other languages; by its purpose, the chart underneath does not rectify the 'IPA is only for English' interpretation either. Our user just learnt a lie. An encyclopedia must prevent such. -- SomeHuman 2006-06-29 18:17 (UTC)
 * By all means disambiguate it if you like, but don't do it to further the ends of a project whose concerns are external to the concerns of writing a good article on the IPA chart for English. Your original stated intentions with this change were to co-opt this article for your and your fellow project members' ends. It sounds like now you have the clarity of the lead at heart instead. Am I right in that? &mdash; Saxifrage ✎ 21:46, 29 June 2006 (UTC)
 * Close enough: that was my primary concern from start. I just found it all the more important because if the project would end up linking to this page, far more people would read it, often without following further IPA links; the suggestion I had made for the project would hardly be convincing if the IPA chart could be misleading. Also – since then some links might, at least till someone corrects them, end up here even when this chart is not the best (foreign sounds) – it would be necessary for the reader to realize he could better follow the link to the general IPA article. The improvement however, is required regardless the page's number of 'hits'. -- SomeHuman 2006-06-30 05:09 (UTC)

American pronunciation of "buoy"
In RP 'buoy' sounds much like 'boy' but in American English it's a diphthong I can't find represented in the table. The sound might occur in other, perhaps even non-American, English words, but my hearing as a non-native speaker of English is not reliable enough to establish such. So, should the table be expanded? -- SomeHuman 2006-06-28 01:57 (UTC)
 * I'd say in American English it's a two-syllable word, homophonous with some pronunciations of Bowie (for me the usual pronounciation of Bowie knife but not of David Bowie which is two-syllable ). User:Angr 07:57, 28 June 2006 (UTC)
 * I'd agree. Most languages with dipthongs move from [-high] to [+high] between the components of a dipthong (consider /au/ or /oi/), with a [+high] to [-high] shift behaving instead like a syllable boundary. AmE is one of these. &mdash; Saxifrage ✎ 08:50, 28 June 2006 (UTC)
 * Yeah, but ['bu.i] is a shift from [+high] to [+high]. And anyway, lots of languages have centering diphthongs. User:Angr 08:54, 28 June 2006 (UTC)
 * Erm, you're right. I shouldn't be deriving linguistics rules from first principles at this time of night, I can see! I believe the rule holds when its [+high] to [-high]... But formulations aside, it's true that /bui/ has two syllables in English. &mdash; Saxifrage ✎ 09:22, 28 June 2006 (UTC)

Broad IPA for international English
There is a discussion to determine a generic, broad application of IPA for international English. If reasonable consensus can be reached, it could become a guideline for writing an IPA pronunciation of English words in wikipedia, without having to worry about the various dialects spoken in different parts of the world. Anyone wanting to comment is welcome at Wikipedia talk:Manual of Style (pronunciation). &minus;Woodstone 14:48, 27 August 2006 (UTC)

"General American" Doesn't Exist
The concept of "General American" dialect is false -- what is described as the geographical range of "General American" (apparently by early and inaccurate, and/or British and inaccurate dialectologists) is the entire Northern dialect group.

There's no reason not to use the term "Standard American" -- the pronunciation is found in authoritative American-published dictionaries, which it appears is as good an authority as that of the Italian dictionaries (and not government bureau) that define the famous and never quibbled-with "Standard Italian". —The preceding unsigned comment was added by 75.6.229.59 (talk • contribs) 11:38, 22 November 2006  (UTC)
 * The term General American is widely used, while the term "Standard American" isn't. As an accent, General American certainly does exist, though as a dialect (i.e. including aspects above and beyond pronunciation, like lexicon, morphology, and syntax) it probably doesn't. And calling the accent "Standard American" would incorrectly imply that it has the status of a standard language, which it doesn't. —Angr 12:34, 22 November 2006 (UTC)

New template idea (crosspost)
I have an idea for making IPA symbols more comprehensible, like this: ʒ (try rolling over that with your mouse). That is,. To discuss the concept, go here: template talk:Ʒ.--Homunq 01:46, 6 December 2006 (UTC)

Audio samples
Hi, what do you think about adding audio samples to the table? Vowels would be a bit tricky, but I think it would work nicely with consonants. Example:

--Kjoonlee 05:05, 13 December 2006 (UTC)
 * Hmm, I guess it wouldn't be that helpful here, because most people are familiar with English sounds. --Kjoonlee 14:19, 24 December 2006 (UTC)

= uh-oh
I added a line for, giving uh-oh as the reference, with a footnote that referred to uh-uh (no), ah-ah-ah (watch it), and kitten (allophone in some dialects). It was promptly reverted with the comment that this is not an English sound. I disagree (What are my examples if they aren't English?), but rather than debate that useless point, I'd like to ask what this chart is for. Is it meant to tell you what sounds are English, or is it meant to help you understand as many IPA sounds as possible using your knowledge of English? I'd prefer the latter, what do others think?--200.6.254.196 14:10, 18 December 2006 (UTC)
 * It's not an English phoneme, which is more specific than saying it's not an English sound. The primary aim of this chart is to help people encountering IPA transcriptions of English words in Wikipedia articles; since only phonemic transcriptions are used, the glottal stop symbol won't be encountered. —Angr 14:24, 18 December 2006 (UTC)
 * It's not exactly the case that only phonemic transcriptions are used. There's no clear instruction on phonemic vs. phonetic in this context, and some pages do use phonetic transcriptions.  I don't know whether any of them use, but List of words of disputed pronunciation uses .  (I don't think it should, but it does.)--JHJ 17:07, 18 December 2006 (UTC)
 * In fact, List of words of disputed pronunciation uses too (Hawaii, which admittedly is a slightly unusual case).--JHJ 17:09, 18 December 2006 (UTC)

As I said, I don't want to argue about whether Hawaii or loch (and thus the phonemes they contain) are English. If this chart is, as you say, in order to help people to understand IPA, and if we give an example of using English words (or, if you don't want to call "uh-oh" a word, then "common, clearly-transcribable utterances of English-speakers"), then it belongs in there. If this chart is intended to be exclusive, then not. Personally, I'd say inclusiveness is a better goal, otherwise this article would be called List of English phonemes, with IPA and examples. en.wikipedia users will indeed encounter the glottal stop in many articles, and should get all the help we can easily give them to understand it. Anyone besides Angr and me have an opinion? --69.79.75.151 01:06, 20 December 2006 (UTC)
 * IMHO limiting this article to the most representative phonemes (not allophones) makes sense. If a transcription uses glottal stops, the articles should link to International Phonetic Alphabet instead. --Kjoonlee 02:50, 20 December 2006 (UTC)


 * I think the main body of the article should be a list of phonemes, more or less as now (though I tend to feel that Scottish English is sufficiently phonologically different from RP and GenAm that it should get a column too if a reasonably standard transcription exists). Below that, though, there could be a separate list of other symbols sometimes found in transcriptions: e.g. ,, and nasalised vowels.--JHJ 12:51, 20 December 2006 (UTC)
 * Yes, I agree a separate list of allophones at the bottom might be helpful. --Kjoonlee 13:54, 20 December 2006 (UTC)
 * I don't. Allophones can be discussed at International Phonetic Alphabet for English, or better yet at English phonology. This chart should be kept short and sweet. —Angr 13:58, 20 December 2006 (UTC)
 * That would be nice too. :) IMHO allophones should be added to both "International Phonetic Alphabet for English" and "English phonology". --Kjoonlee 04:04, 21 December 2006 (UTC)

This is not (just) an allophone! Uh-uh vs. uh-huh is a minimal pair, and it exists in all three main dialects discussed here! The reason to keep allophones out of here is to avoid multiple transcriptions for words, and I understand that, but "uh-oh" is not going to lead to that, and besides, we should remember that this page serves as a decoding reference at least 5 times as often as it serves as an encoding reference. Hawai'i is a good example of an English word plenty of people have to look up how to say, and there's simply no better way to help them than to include  (and besides, there's an empty space on the page, fer cryin out loud). IS an allophone, there are no minimal pairs involving it in any English dialect; but going by number-of-speakers (though not by antiquity), is more of a phoneme than. --Homunq 00:34, 9 January 2007 (UTC)


 * Since this chart will be used by people who want to decode IPA transcriptions they see on wikipedia, I think a list of "special symbols used in transcribing some English words" would be appropriate at the bottom of the page. Most dictionaries have such a table in their pronunciation guides (for nasalized vowels, borrowed sounds like /x/ and sounds like /?/ that nobody knows what to do with). Since these sounds are (peripherally, at least) part of English phonology, and since they might trip people up on pages, I think they should be included here.--Gheuf 17:59, 25 January 2007 (UTC)

Happy
I've been of accused of providing original research, but this is not true. My changes were made on good faith after looking through a few dictionaries and websites dealing with phonetics. The dictionaries that I have looked at — both hard copy and online — generally transcribe 'happy' as  (not  ) such as Oxford, Chambers (using their own pronunciation spellings, ie, "hap'i", where 'i' is short as in 'sit') and Collins. The Macquarie and Merriam-Webster dictionaries on the other hand transcribe 'happy' as  and "'ha-pE" respectively, but note they also transcribe 'creep' as  and "'krEp" respectively. Now this indicates that the 'y' in 'happy&#39; and the 'ee' in 'creep' are the same at the phonemic level, but could be different at the phonetic level. If the [i] is short in happy then either it should be spelt  or , but because this page is dealing with phonemic transcriptions rather than phonetic transcriptions then either   or   should be used, but not. Marco 15:33, 24 December 2006 (UTC)


 * It find it rather strange to go the phonetic level for &lt;i&gt; haggling over the difference between and, while ignoring completely that  is surely not , but . So a more correct rendering would be . &minus;Woodstone 16:05, 24 December 2006 (UTC)


 * I think I'll have to point some people to Happy tensing. --Kjoonlee 16:09, 24 December 2006 (UTC)


 * Take a look at Longman, where creep is and happy is . American dictionaries like MW aren't the best for this because they don't show phonemic vowel length anyway. —Angr 17:40, 24 December 2006 (UTC)


 * I think Oxford's current transcription (for RPish accents) is actually /hapi/, with the same use of /i/ as in Longman (apparently meaning "some people think this is and some think it is ").  Does anyone actually think it's a separate phoneme?  (NB, given the previous comments: the /a/ in that transcription is not an error.)--JHJ 21:55, 26 December 2006 (UTC)

Thankyou Woodstone for correcting me, the use of /a/ instead of /æ/ was an oversight, but my focus is on /iː/. And he is correct that is a correct rendering, but only for RP;  is the correct rendering for Australian English; and  is probably the correct rendering for General American according to Angr. So I think a better example should be found. I will also mention that the Oxford Australian English Dictionary renders 'happy' as, showing phonemic lengths that Macquarie leaves out. Angr mentioned one dictionary, which according to him uses phonetic transcriptions rather than phonemic transcriptions, which is unusual as most I know of use phonemic transcriptions. I know many American dictionaries such as MW aren't the best but their own systems inply that the lengths are there through examples. Chambers does this too. Have a look at some examples by Australian linguists from Macquarie University of Australian phonemic and phonetic renderings of words with a similar final vowel as in 'happy&#39;. So can a better example be found that is phonemically similar in all three dialects? Marco 10:59, 25 December 2006 (UTC)
 * I have no idea about accents, but would "Java" do? --Kjoonlee 16:39, 25 December 2006 (UTC)
 * Do for what? It doesn't have the "happy" vowel at all. —Angr 16:56, 25 December 2006 (UTC)
 * I thought Marco asked if we could use an entirely different word that can be used to illustrate primary stress. From the article, it looks like the first syllable of Java is not the same in AuE, so I guess we'd have to find something else. --Kjoonlee 17:01, 25 December 2006 (UTC)
 * Tapping? It has no diacritics or exotic symbols other than the stress mark. --Kjoonlee 17:13, 25 December 2006 (UTC)

Batting
We need a better example for syllable separation because CVCVC is pretty much unambiguous (CV CVC or CVC VC). "Sighing" was CVGVC/CVVVC, making syllable separators more relevant. --Kjoonlee 16:06, 24 December 2006 (UTC)
 * The previous example, sighing was much better, as the syllable break in that word is unambiguous. "Batting" is actually ambiguous because of the hypothesis of ambisyllabicity, according to which the [t] (or  in accents with flapping) is simultaneously the coda of the first syllable and the onset of the second. —Angr 17:37, 24 December 2006 (UTC)


 * I was trying to find a word that was dialect neutral because 'sighing' is a bad example for Australian English wrt phonemic spelling, just like 'happy' is. Thankyou Angr for explaining the situation of /t/ in words such as "batting". I think some speakers in Australian English have this flapping. Can a better word be found that is phonemically similar in all three dialects? Marco 11:02, 25 December 2006 (UTC)


 * How about plankton? I believe it's in RP, GenAm, and GenAus. —Angr 12:23, 25 December 2006 (UTC)


 * That's a much better example than the current one. So no objections to its use. Marco 12:28, 25 December 2006 (UTC)