Talk:International Phonetic Alphabet/Archive 5

Original research

The section discussing the advantages and disadvantages of using the IPA in dictionaries looks like original research to me. If published research has been done on the issue, please cite it. Otherwise, please remove it. Angr 14:57, 5 October 2006 (UTC)

I left a note on the talkpage of the person who added it in. They haven't responded yet, so I guess the removal is a good idea. We can always pull it out of history if it does get sourced. The ikiroid (talk·desk·Advise me) 19:41, 5 October 2006 (UTC)

Information Removed

I removed this information. Some of it is questionably notable, all of it is unsourced, most of it could easily be sourced by referring to dictionaries which are described here. The ikiroid (talk·desk·Advise me) 20:17, 8 October 2006 (UTC)

On the other hand, the systems used in American dictionaries have their drawbacks. One of these is the difficulty of properly representing sounds not found in the English language, such as [r] and [ɾ] (though some solve this problem by having special symbols for representing foreign sounds). Another, more significant one is that each dictionary uses its own system of phonetic notation, and hence there is little consistency among different dictionaries.

The IPA is also not universal among dictionaries in other countries and languages. Mass-market Czech multilingual dictionaries, for instance, tend to use the IPA only for sounds not found in the Czech language, e.g. ð, θ, ŋ, æ, ə, etc. It is obvious that the phonetic transcription uses the Czech alphabet (e.g. š instead of ʃ, č instead of tʃ, etc. – cheese [či:z]). Some symbols are replaced by letters that denote close sounds in Czech, e.g. home [houm] instead of [həʊm]. However, better dictionaries give some comments on the differences in pronuciation in English and Czech. The transcription described above is used because it is more understandable for Czech users than more precise IPA characters.

I've restored one sentence from the Czech paragraph for now, since the information is not in dispute. I think it would look silly to cite, for example, 10 Czech multilingual dictionaries to prove this point. -- Mwalcoff 21:16, 8 October 2006 (UTC)

Yes, I am a native Czech speaker but I don't know any Czech multilingual dictionary which uses IPA transcription. IPA is neither used in Czech linguistic literature. If so, it is a very special and rare case. The transcription described above results from a longtime tradition in the Czech lexicography. --Pajast 12:42, 9 October 2006 (UTC)

Can you at least cite any dictionary that has this as an example? It would help. The ikiroid (talk·desk·Advise me) 15:32, 9 October 2006 (UTC)

If I don't forget, I'll bring you some examples tomorrow. It is any dictionary which has "anglicko-český slovník" in its name. --Pajast 15:58, 9 October 2006 (UTC)

The version I have is the very popular (if not very good) KPS pocket-size Anglicko-Český / Česko-Anglický Slovník Gramatika Fráze. It only uses IPA for sounds not in the Czech language. For example, it transcribes "shatter" as "šætə," combining the Czech symbol for "sh" with the IPA symbols for what we call "short a" and "schwa." -- Mwalcoff 16:10, 9 October 2006 (UTC)

And this is what (almost?) all Czech dictionaries and textbooks do. --Pajast 06:56, 10 October 2006 (UTC)

As I promised, I give some examples of the dictionaries which use modified IPA (books I've got in my bookcase):

• Hais K., Hodek B. Velký anglicko-český slovník. Academia, Praha

I. A-E, 1991. ISBN 80-200-0065-8.

II. F-M, 1992. ISBN 80-200-0066-6.

III. N-S, 1992. ISBN 80-200-0067-4.

IV. T-Z, 1993. ISBN 80-200-0478-5.

• Fronek J. Velký anglicko-český slovník. Leda, Praha 2006. ISBN 80-7335-022-X.
• Paroubková J. et al. Lékařský Slovník anglicko-český a česko-anglický. Avicenum, Praha 1991. ISBN 80-201-0178-0.
• Kernerman L. Vácha J. Password – Anglický výkladový slovník s českými ekvivalenty. Mladá Fronta, Praha 1991. ISBN 80-204-0288-8.

An example of Czech liguistic literature (Czech grammar - chapters on the Czech phonetics, phonology and pronunciation):

• Karlík P., Nekula M., Rusínová Z. (eds.) Příruční mluvnice češtiny. Nakladatelství Lidové noviny, Praha 1995. ISBN 80-7106-134-4.

Josef Fronek in his dictionary wrote:

"In accordace with long-established Czech lexicographical tradition, a modified version of the International Phonetic Alphabet (IPA) is adopted in which letters of the Czech alphabet are employed. They are used as follows:" (The list of characters and examples of their use follow.)

This is an example of the only book I've found which is written in the Czech language and in which IPA is used in a proper way:

• Duběda T. Jazyky a jejich zvuky. Univerzálie a typologie ve fonetice a fonologii. Karolinum, Praha 2005. ISBN 80-246-1073-6.

It is a phonological publication which compares various languages spoken in the world.

--Pajast 06:56, 10 October 2006 (UTC)

I'm sourcing the article right as a write this. The ikiroid (talk·desk·Advise me) 22:12, 10 October 2006 (UTC)

I was in a large bookshop yesterday. I was browsing through a lot of various dictionaries. I must say that I have found some "modern" dictionaries and textbooks that used more IPA letters (but not completely all in some cases) than others do. They break down the long tradition. But the major part of the available publications keeps traditional transcription. I feel it is user-friendly because I think that few Czech students understand the IPA well. The advantage of the Czech orthographical system is that it enables using Czech letters for transcription instead of (some) IPA characters. Yes, the IPA is universal wordlwide, neither the Czech alphabet is. But it is a better choice for the average Czech user. --Pajast 14:20, 11 October 2006 (UTC)

Wikipedia needs a robot voice pronuciation generator

I have no idea how to read this IPA funny symbol thing after words. What wikipedia needs is a computer generated MP3. Some people need to campaign hardcore for this feature. Honestly it could be as simple as finding a text to speech bot online that supports these IPA symbols and inserting the word into some URL like bot.com/query?meme%20funnysymboldsgft so wikipedia people click the pronunciation and download an MP3. But it would have to be easy to use for wikipedia people so make some linky keyboard that generates the appropriate mp3 link.

You gotta admit this is a pretty good idea for an online encyclopedia. Paste this to the appropriate wikipedia feature request page if you can, I don't know wikipedia very well, im new to this. —Preceding unsigned comment added by 64.40.45.117 (talk • contribs) 18:40, 8 October 2006

Unfortunately, to the best of our knowledge no system that supports this functionality exists, and to develop one would require a very large research and development investment. Nohat 20:43, 8 October 2006 (UTC)

Or, we do it the way that Wikipedia was made in the first place, and recruit volunteers to speak every word and phrase that has a Wikipedia article associated with it. For a million articles we'd just need 10,000 editors to read 100 words (or titles) each that they know how to pronounce. This is the approach Microsoft has taken for years with their CD-ROM dictionaries. The idea is great, mostly because, statistically speaking, nobody uses IPA, and nobody understands it, and the IPA notations in articles are in fact useless to our readership. Tempshill 16:55, 3 November 2006 (UTC)

Rather, it's useless to that large part of our readership that can't be bothered to acquaint itself with the IPA. I agree that including an MP3 file is a great idea, but to say that:

statistically speaking, nobody uses IPA, and nobody understands it,

strikes me as a somewhat disigenuous use of statistics. After all, the IPA doesn't seem to bother the users of most quality dictionaries in the UK. I have no problem at all with providing supplementary information on pronunciation - I agree that it should be included -, but the generally anti-IPA attitudes on these pages, which seem to imply removing the information altogether, trouble me. It feels rather too much like a concession to ignorance. Just to learn the basics of the IPA (and it's really not that hard) is very worthwhile. You not only come into possession of a system that helps you understand pronunciation guides, but you may even gain a better understanding of how speech works. It's like basic maths, or basic geography, or a basic knowledge of the history of your country. The benefits are well worth the (relatively small amount of) trouble. garik 13:03, 26 November 2006 (UTC)

GA review

This article has several {{Fact}} tags, and as such cannot pass GA as of yet, since GA requires all content to be accurate and verifiable. Otherwise, it seems a strong article. Adam Cuerden ^talk 09:32, 12 October 2006 (UTC)

OK, it only has three, and two of them are about the same thing. In all, we need a reference regarding the 1993 revision of removing implosives, and then we need a reference regarding the connection between [ʕ] and [ع]. Does anyone have any sources for these two things? I'm all out. The ikiroid (talk·desk·Advise me) 21:53, 12 October 2006 (UTC)

Building off of Adam Cuerden's point, there are also a few sections which do not have inline citations to any references, and are unmarked as needing citation. Each section of the article should have at least two inline citations, since the content does need to be verifiable. Thank you for taking the time to address these concerns; the article does look quite good overall. -Fsotrain⁰⁹ 03:41, 14 October 2006 (UTC)

"hoş geldiniz" in IPA?

Hi, can u pls give me the turkish "hoş geldiniz" in ipa alphabet? thanks feoe@campus.upb.de --Foerdi 20:15, 14 October 2006 (UTC)

[hoʃ geldiniz], [hoʃ ɟeldiniz], or [hoʃ ɟeld̪in̪iz], depending on how detailed you want to be. The second transcription followed the conventions of the Turkish language article. --SameerKhan 20:50, 14 October 2006 (UTC)

I added the { { IPA | } } template above to read the IPA characters. Hope you dont mind. :) --Haldrik 22:51, 12 November 2006 (UTC)

Four?

Hi, I was reading this article and noticed that there seems to be something wrong with this sentence:

The symbols of the International Phonetic Alphabet are divided into four categories: letters, diacritics, and suprasegmentals (symbols that indicate such things as the tone and inflection of a spoken utterance).

Only 3 categories are listed. Am I missing something? This is my first time thinking about editing Wikipedia. 206.154.229.138 14:27, 16 October 2006 (UTC)Patrick

GA Passed

This artcle has passed the GA nominations. As a few suggestions I would suggest converting the list in the "Description" section into prose ans then try seeking a peer review. --Tarret 22:46, 16 October 2006 (UTC)

I've done the conversion, please copy-edit and/or rephrase as appropriate. Thank you Tarret. -Fsotrain⁰⁹ 22:58, 16 October 2006 (UTC)

Use on Wikipedia

Why is this being spread on Wikipedia rather than using analogous words which are more understandable (and readable) to most readers? —Centrx→talk • 21:02, 18 October 2006 (UTC)

Because the IPA is precise and unambiguous, and sound-alike transcription is not. It's true that an encyclopedia needs to be as accessible as possible, but it should not compromise precision in favor of accessibility. Anyway, there's always the IPA chart for English if you need help reading IPA transcriptions. (Yes, it has been argued that one shouldn't have to learn the IPA in order to find out the pronunciations of words, but that's like saying that one shouldn't have to learn calculus in order to understand what a Hilbert space is.) --Siva 21:27, 18 October 2006 (UTC)

Not a very good analogy. Hilbert Space can only been learned through calculus, proper pronunciation of a word can be learned in any number of ways. I think IPA is a good thing, but that including informal "pidgin" pronunciation spellings is also acceptable and much more accesible to newcomers. From the Dorothea Lange article, most people are NOT going to know what to make of "/ˌdor.ə.ˈθi.ə.leŋ/" and shouldn't be expected to. "dor-a-THEE-a LANG" is close enough for common usage. - Richfife 06:24, 23 October 2006 (UTC)

Agree that while IPA is great pidgen shouldn't be completely outlawed. -Ravedave ^{(help name my baby)} 07:01, 23 October 2006 (UTC)

I thought of a better analogy: Classical mechanics vs. String Theory. Classical Mechanics was disproven 100 years ago and is now known to never accurately model the physical world, and yet it is still used 99.99% of the time for almost all purposes. Why? Because it does the work and is simple enough to be used day by day. The same goes with informally spelled pronunciations. - Richfife 16:15, 23 October 2006 (UTC)

Then it's a very poor analogy because ad-hoc pronunciation guides do not do the work and they aren't any simpler than the IPA. —Angr 16:35, 23 October 2006 (UTC)

Once someone has memorized the IPA table, then yes, IPA is just as simple. 99% of Wikipedia users have not (don't kid yourself). Sending them to the table to crossreference each phoneme isn't practical. And, as I said, I'm not advocating removal of IPA, only allowing the inclusion of ad-hoc guides. The average user does not care about the fine points of different kinds of stress. The only wish to pronounce names more or less correctly in conversation so they are understood. If they wish to spend the time to be absolutely correct, then the IPA guide is sitting right there. - Richfife 17:22, 23 October 2006 (UTC)

But learning an ad-hoc guide requires just as much effort as learning the IPA, with the disadvantage that the ad-hoc guide you've learned only works within the one publication that's using it and for the one language it's using it for (and in some cases not even for that). It's like paying $100 for poorly designed software that only works on one interface (when it works at all), when for $100 you could get well designed and highly reliable software that can be used on any platform. —Angr 17:32, 23 October 2006 (UTC)

I disagree. I'm not advocating creating a separate ad-hoc guide on another page that people should reference, I'm advocating using spellings that specifically use the characters to represent their most common pronunciations. Anyone that looks at "th" or "k" and doesn't know the sound they represent is deliberately being obtuse. They approach is imprecise, yes, but it's also intuitive and in situations where it can't convey the pronunciation well, it shouldn't be used at all. Also, even though the focus of the content of the English Wikipedia is not on the English speaking world, it is assumed that the readers are familiar with English, so having English specific pronunciation keys is OK. - Richfife 17:52, 23 October 2006 (UTC)

Yes, the pronunciation of "k" is pretty obvious, which is why it's represented [k] in the IPA. The sound represented by "th" isn't obvious at all, since that combination of letters is used to spell two different sounds in English (which is why tooth and smooth don't rhyme, although they look like they should). Also, since there are so many different varieties of English, using an ad-hoc key based on English spelling raises the question of whose English we're using. If an English or Australian editor says that Föhn is pronounced like fern, that's close enough, but if a Scottish or American reader sees that, they're not going to get anywhere close to the right pronunciation. —Angr 18:18, 23 October 2006 (UTC)

- :: (Reply to post just above. Pulling back indents a little). As I mentioned above, if the informal approach doesn't work, it shouldn't be used. I disagree on the "whose English" point as well. I watch BBC News in the US almost every day. A British accent is certainly identifiable, but the differentiation of vowel sounds, etc. is grossly exaggerated. - Richfife 21:17, 24 October 2006 (UTC)

I have made the link to the International_Phonetic_Alphabet_for_English more visible in the page header. I suggest most of the traffic to this page is people wanting to see how to pronounce a word which has been sounded out in IPA in a Wikipedia article. I suggest many like myself are then baffled by what they find. Ideally all links from "IPA" next to a sounded word should link to IPA chart for English in the first instance. Lumos3 14:02, 19 October 2006 (UTC)

For that purpose, IPA chart for English is probably even better. But only, of course, if the word itself is English. I frequently add pronunciation guides for articles whose titles are non-English words, in which case International Phonetic Alphabet for English and IPA chart for English aren't any use at all. —Angr 14:14, 19 October 2006 (UTC)

Surely the IPA chart for English explains how to pronounce the IPA sounds with reference to familiar English pronunciation of words. A foreign word would be rendered in just one way in IPA and speakers of different languages would then go to their own language chart to see how it’s sounds are pronounced by referring to words that are familiar to them. Since The English Wikipedia is being read by English speakers then the IPA chart for English is the only one needed. Am I missing something here? Lumos3 12:29, 20 October 2006 (UTC)

The IPA chart for English only discusses sounds used in English. When I add the German pronunciation of a German name or the Irish pronunciation of an Irish name, it's going to include symbols that aren't mentioned at IPA chart for English. Thus instead of linking there, a link to this page is provided. (However, it may actually be more helpful to link to German phonology and Irish phonology respectively, instead.) —Angr 12:51, 20 October 2006 (UTC)

See also Wikipedia talk:Manual of Style (pronunciation), where there has been recent discussion about standardizing a simple phonemic IPA scheme for English pronunciation in Wikipedia. —Michael Z. 2006-10-20 19:40 Z

The original poster in this thread is right. I would like to remind the IPA boosters here that, statistically speaking, nobody understands IPA and nobody uses it, and it will always be in fact useless to nearly our entire readership. The soundalike scheme would be less precise, yes, but it would actually be used by people. 16:58, 3 November 2006 (UTC) —The preceding unsigned comment was added by Tempshill (talk • contribs) .

Which statistics are you referring to? It would be nice to read a peer-reviewed study about the effectiveness of IPA and sound-alike respelling schemes, to put to rest all of the baseless claims about one being better or more widely understood than the other. —Michael Z. 2006-11-03 20:28 Z

First of all, please sign your posts with your name, not just the time. Secondly, no, in fact, it wouldn't. The "soundalike schemes" are not just imprecise, they're just as difficult to learn as the IPA, and therefore worse than useless. —Angr 17:27, 3 November 2006 (UTC)

How are they difficult to learn? The elements of a sound-alike scheme are simple syllable formations that people already know. There is no learning required by any native English speaker. Most Wikipedia readers are not going to learn and remember IPA, no matter how many articles refer to it; if pronunciation information is included in articles, the sound-alike scheme needs also be included, which you can find commonly in the BBC and the New York Times ("Worse than useless"?). —Centrx→talk • 17:57, 3 November 2006 (UTC)

Centrx, you say 'There is no learning required by any native English speaker'. However, this edition of Wikipedia is international in scope and everyone knows that it is not just used and read by native speakers of English alone. For the part of our readers that does not speak English natively, IPA is a much better transcription system than sound-alike schemes, because the interpretation of the latter depends on the reader's idea of English pronunciation, while IPA is universal. (Note that I'm not talking about the reader's overall competence in English', to avoid the 'if they don't speak it, they won't read it either' straw man.) — mark ✎ 19:06, 3 November 2006 (UTC)

Indeed, and even for the readers who are native speakers, ad-hoc systems are worse than useless when it comes to the pronunciation of non-English words. We have articles on people and places from all over the globe, and no "sounds like/rhymes with" system is going to tell us how to pronounce the name of the town in Hungary or the Vietnamese politician whose article we're reading. IPA is indispensable for that. —Angr 20:27, 3 November 2006 (UTC)

The above two entries are ivory-tower scholar talk at its worst. I concede gladly that IPA is a "much better transcription system" for the purpose of accuracy. But I maintain — and I see neither of you have addressed this point — that, statistically speaking, nobody understands IPA and nobody uses it. It is obvious that for the layman who is curious how to pronounce a word, a soundalike scheme is twelve billion times superior to an IPA transcription, because the soundalike scheme will be used. The fact it's less accurate pales compared to the fact that IPA is not usable and will never be usable by the layman. And after all let's keep in mind that this is not a brain surgery manual where lives hang in the balance. We're presenting information here that is intended to be readable. We want people who are not linguists to be able to come pretty close to pronouncing words — that's it. Tempshill 20:32, 3 November 2006 (UTC)

This is an encyclopedia. Encyclopedias are academic works. Encyclopedias do not deny their readers accurate information for fear their readers might not understand it. Readers who are unwilling to spend the 30-45 minutes necessary to learn how to read the IPA can simply skip over the pronunciation section of articles. And your repeated claims that "statistically speaking, nobody understands IPA and nobody uses it" are anti-intellectualism at its worst and are also highly insulting to the hundreds of thousands of people who have spent the 30-45 minutes required to learn to read IPA. —Angr 21:04, 3 November 2006 (UTC)

You are going to have to cite a source if you are making the remarkable assertion that hundreds of thousands of people can read IPA. I believe your perception of reality in this area has been distorted by your Ph.D in theoretical linguistics, and would gladly bet you ten bucks that 100% of the undergrad students you have (presumably) taught IPA have forgotten all of it by now.

More to the point, encyclopedias are concise academic works and are not exhaustive. Your statement is untrue about encyclopedias "denying" their readers accurate information — by definition we do that in every article. I believe that, because IPA is used by linguists only, it's useless, overly-technical jargon that actually detracts from the article, just like garden-variety overly-technical jargon detracts from Wikipedia articles. Tempshill 17:45, 4 November 2006 (UTC)

I would say that general-knowledge encyclopedias are not academic works but are meant to be accessible to the masses. I couldn't imagine World Book or Encarta using IPA. Incidentally, I've been using Wikipedia for a year, and I still have to spend time figuring out the difference between /ɑ/, /a/ and /ɒ/. -- Mwalcoff 00:42, 5 November 2006 (UTC)

So, in other words, with sound-alike, non-native speakers may have to learn it, and with IPA everyone has to learn it? —Centrx→talk • 23:50, 3 November 2006 (UTC)

Note that it may or may not be fine to have this in articles, but as something separate—it should not be the second word of every article—and not at the expense of a sound-alike scheme. —Centrx→talk • 23:52, 3 November 2006 (UTC)

Can someone please create a layman's version of this page because it confuses the living #### out of me

thanks. Youaredj 22:12, 31 October 2006 (UTC)

Have you seen IPA chart for English? It only covers the sounds of English, not other languages, but it keeps technical terms to a minimum. —Angr 06:30, 1 November 2006 (UTC)

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, {{Ʒ}}. To discuss the concept, go here: template talk:Ʒ.--Homunq 01:47, 6 December 2006 (UTC)

I think I understand your idea: that each individual IPA character would link to a page giving a clear explanation of how it's pronounced in plain English? I think that's an excellent idea. garik 11:56, 6 December 2006 (UTC)

Thanks for the comment, I've moved it to the main discussion on the talk page mentioned above, and

responded there.--200.6.254.170 20:49, 6 December 2006 (UTC)

Have you? Now I look at this again, I see I didn't quite understand anyway. Oh well... garik 21:00, 6 December 2006 (UTC)

IPA and Unicode

Hi, Unicode 5.0 is now out. Does anyone know if the labiodental flap has been incorporated into the latest version of the Unicode standard? Thank you. --Kjoonlee 18:08, 6 December 2006 (UTC)

I'm pretty sure it hasn't. —Angr 18:35, 6 December 2006 (UTC)

Thanks. :) SIL Corporate PUA Assignments say it's to be included in a later version, and Proposed New Characters: Pipeline Table mentions it's still in the pipeline. --Kjoonlee 16:24, 7 December 2006 (UTC)