Talk:Phonemic orthography

From Wikipedia, the free encyclopedia

Contents 1 True alphabets 2 International Phonetic Alphabet 3 Rename to: phonemic and phonetic orthography 4 Interlingua 5 cot/caught merger 6 Spanish 7 Wrong example 8 grapheme-to-phoneme correspondence

[edit] True alphabets

Georgian language claims that there are 14 true alphabets. Anyone know what they are?

Sounds like a nonsense claim to me... Morwen 23:03, 25 Nov 2003 (UTC)

[edit] International Phonetic Alphabet

How can International Phonetic Alphabet be used as an example of a phonemic alphabet? Don't you see the phonemic x phonetic conflict mentioned later in the article?

You can do this by using IPA characters in order to write the phonemes of the language. Roman orthography is practically useless for a lot of languages, and the IPA is a good starting point for a phonemic orthography. Sure, it's primarily a phonetic alphabet, but that doesn't mean it can't be used as a phonemic alphabet by omitting some features. thefamouseccles

Quite. The requirements of a phonemic orthography are basically a subset of those for a phonetic orthography . -- Q Chris (talk) 15:07, 23 November 2007 (UTC)

[edit] Rename to: phonemic and phonetic orthography

The article seems to talk equally much on both topics (phonemic and phonetic orthographies). Perhaps rename (and restructure) it to cover both notions?--Imz 06:41, 22 November 2005 (UTC)

[edit] Interlingua

How can Interlingua be considered phonemic when the sound [f] can be spelled "f" or "ph", the sound [r] "r" or "rh", the sound [t] "t" or "th", the "c" can be read [k] or [ts], the "ch" can be read [k] or [ʃ], the [k] can be spelled "c", "q" or "ch", the "g" can be read [g] or [ʒ], the "j" can be read [ʒ] or [j], the [ts] can be spelled "c" or "t", the "w" can be read [w] or [v]?

That characterization of Interlingua is highly misleading. It includes pronunciations that occur only in foreign words ("w" as [w] or [v]), pronunciations that are not generally accepted ("j" as [j]), and repeats one complaint three time in different variations ("ch" as [k] or [ʃ], [k] as "c", "q" or "ch", [ts] as "c" or "t"). Interlingua has a few exceptions to phonemicity, but so does Finnish, which is listed as having good phonemic correspondence.

Here is a more accurate representation of Interlingua's letters and sounds. There are also the digraphs ph, th, and rh, but they have one sound as well.

a one sound

b one sound

c two sounds

d one sound

e one sound

f one sound

g two sounds

h one sound

i basically one sound, can be a semi-consonant

j one sound

k one sound

l one sound

m one sound

n one sound

o one sound

p one sound

q one sound

r one sound

s basically one sound

t marginally two sounds

u basically one sound, can be a semi-consonant

v one sound

w rare

x basically one sound

y one sound

z one sound

I listed Interlingua as having "a good grapheme-to-phoneme correspondence", not as being phonemic. 66.68.174.245 05:41, 18 February 2007 (UTC)

You realize of course that you have weakened the definition of phonemic orthography to make it just mean predictable pronunciation based on spelling without also implying the opposite. Now one can claim that French has a phonemic orthography, which I don't think is the intention. -- Dissident (Talk) 01:33, 20 February 2007 (UTC)

I think there are different terms for languages where there is a predictable pronunciation based on spelling (e.g. French) and languages where there is a predictable spelling based on sound (I believe Arabic is like this). I can't remember what they are though. Does anyone know? -- Q Chris 08:13, 20 February 2007 (UTC)

I removed Interlingua, it is a bad example of phonemic orthography.Cameron Nedland 21:06, 15 April 2007 (UTC)

[edit] cot/caught merger

For an example of the difficulty in creating a phonemic alphabet for English, the cot/caught merger is probably a better example. Bad/lad split seems to affect less than 10 words, but the caught/cot merger affects a huge chunk of the English language, and is also a lot more recognizable to language enthusiasts. —Preceding unsigned comment added by 71.212.27.154 (talk) 00:53, 14 October 2007 (UTC)

[edit] Spanish

I don't feel confident enough to edit the article directly without discussion, but as a native Spanish speaker (from Spain, not Latin-America) I'm under the impression that Spanish has more than a good "grapheme-to-phoneme" correspondence, and yet it's not included in the list, which is pretty stricking to me, mostly considering other more minoritary languages, such as Basque (euskara), Finnish, Sanskrit, or even constructed languages that very little people speak at all.

I didn't edit the article myself because I can't prove Spanish would be comparable to them in "grapheme-to-phoneme" terms with more than my own experience as a native speaker, but I would like someone to tell me about this and justify, if possible, why it is not included as I believe it should be.

Yeah, I think it has a pretty good correspondance too. The only reason I would think that it didn't would be because of silent h, the variation in usage of x (sometimes it's /x/, sometimes it's /ks/), and the letters c and z which both represent the same thing. In yeismo dialects ll and y also cover the same phonemic territory and seseo dialects merge the sound of c/z with s.

The neat thing about Spanish is that when you see a word spelled, you know exactly how it's pronounced. It's a little different the other way around--that is, when you hear it you may not spell it correctly-- but instances of the latter are not very common. If Spanish doesn't fit into the category, this sort of thing, I think, can be put into the article. Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 18:17, 27 October 2007 (UTC)

I wouldn't consider yeismo and seseo (and ceceo in certain parts of Andalucia) real dialects but more like divergent accents. An educated person, even if he speaks with such accents, is aware of which is the correct and canonical pronunciation, which is what matters here, is it?

Of course, there are a few divergences that ARE part of the official orthography, namely:

H is mute when it's written alone and has a sound when accompained of C (CH).

C sounds like Z when in front of E and I (CE, CI) and like K when in fron of A, O and U (CA, CO, CU). Z always sounds the same, but it's not commonly used with E or I, where the use of C is the norm.

Then again, to have the sound of K with E and I, you must use QUE and QUI, and here U becomes mute. KE and KI also exist but...

K is an imported character used mostly for imported foreign words and in most cases it is considered correct to replace it with QU (like in "kilo", that can also be written "quilo", which considered more "native" to Spanish). This does not apply for purely foreing words, which are not regulated by Spanish orthography, of course xD.

The rest of the differences are due to regional accents, and while socially recognized and accepted (and, of course, studied), they are by no means CORRECT according to the canon, so I don't think they should be taken into account for this matter. But I'm not aware of the criterion that is followed, so I may well be wrong. —Preceding unsigned comment added by 213.60.137.253 (talk) 22:40, 27 October 2007 (UTC)

While I'm not so dismissive of nonstandard dialects, your emphasis on standard pronunciation is pretty appropriate in this case, I think. I'd forgotten about c/k/qu... that makes it even less phonemic. I guess that despite Spanish's predictable spelling system, it is not really "phonemic." Heck, Irish and French have predictable spelling but they're not considered phonemic. Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 22:47, 27 October 2007 (UTC)

I guess you're right then, but still, I would like to know what makes the listed languages so much more phonemic in comparison to Castilian (the "original" Spanish, if you please xD). Apart from the listed "inconsistencies" the rest of the characters are allways pronounced the same way, regardless of anything else, with a 1:1 character-to-sound correpondence.

So, how much more phonemic are the other languages listed?

The Devanagari script was designed to represent the sounds of Sanskrit exactly. Prior to the existence of a writing system Mantras were passed on orally, with great care being taken to ensure the sound was reproduced exactly. The Devanagari script (and other Bhrami scripts) were designed to do the same. Pronunciation was scientifically studied in the [Taittiriya Upanishad], which meant that devanagri is not only a Phonemic orthography for Sanskrit, but the also the alphabetical order corresponds to the mouth shape and position of the tongue in the mouth, you can feel you tongue move forward as you produce the consonants of each class in order. Devanagari is also used for other languages such as Hindi, but it is not quite a Phonemic orthography (though it is pretty good) because of sound convergences and sounds introduced from languages other than sanskrit which do not have unique representations. -- Q Chris 07:45, 29 October 2007 (UTC)

It's interesting that Devanagari script, thanks for the info!

Anyway, I just remembered yet another Spanish orthographic rule that concerns character R, which has two sounds, one being like Japanese's RA RI RU RE RO, which is "soft", and another stronger sound.

-When writen at the start of the word it has the "strong" sound, allways.

-And "soft" when written in the middle of a word, except when preceded by N.

-Then again, if you need to achieve the "strong" sound in the middle of a word, you need to use the digraph "RR".

I feel rather stupid for having forgotten this one xD. Now, definitelly, there are no more.

Of course this makes Spanish haven an even less phonetic orthography. When I started this I was focused on the aspect that vowels never change their sounds or combine to create new ones (except for the use of QU), and apparently forgot a few other aspects that mainly concern some consonants. Ah well... It's all clear now.

[edit] Wrong example

Phonemicity may be preserved either by nativizing the loanword's pronunciation (for example, ski is pronounced [ski] in Italian, rather than [ʃi] as in the original Norwegian word)

This example is completely wrong, the word is pronounced [ʃi] in Italian, and for this reason it has seen its spelling changed from ski ("k" isn't used in Italian anyway) to sci ("sc" is how we usually spell the sound [ʃ]). Lupo Azzurro (talk) 16:21, 18 April 2008 (UTC)

The English pronunciation is the result of spelling pronunciation, so it doesn't illustrate the point. I believe there's a better example mentioned at Russian phonology:

"шофёр (from French chauffeur) was pronounced [ʂoˈfɛr] in the early twentieth century but is now pronounced [ʂɐˈfʲor]."

I'll change the article accordingly. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 19:30, 18 April 2008 (UTC)

Sorry, but I'm not a big fan of your fascism example. <sc> is [ʃ] in a number of English words, notably a bunch ending in -science and its relatives, as well as a number of Italian loanwords such as crescendo. Certainly this isn't the pronunciation one ordinarily expects for <sc>, but there are so many cases of it that I don't think this is the best example of irregular spelling. (But if we do keep this example, then [sk] should be changed to just [s], as the latter is much more common before <i>; and the wording needs to be rephrased a bit so as not to imply that fascism is an Italian word: the Italian etymon is fascismo.) —Ruakh_TALK 23:23, 19 April 2008 (UTC)

That's a good point. I guess we can't really make a good case for spellings from most loanwords... into English. How about fajita? — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 23:55, 19 April 2008 (UTC)

Yeah, I think fajita works better. :-) —Ruakh_TALK 01:25, 20 April 2008 (UTC)

[edit] grapheme-to-phoneme correspondence

It says grapheme-to-phoneme correspondence, which is the usual when talking about the very correspondence. However, it is trully a grapheme-to-(allophone or phoneme) correspondence, isn't it? Take this example (which is from the allophone page): cat v table - those t's are different phonemes, but the same allophone. Because the everyday description is going to be this more common name grapheme-to-phoneme, I think we should keep it; but we should also link out to a separate page that defines it, or if not then at least handle the ambiguity here where it occurs. — robbiemuffin ^{page talk} 13:28, 21 May 2008 (UTC)

You've got it backwards. the cat/table example shows the same phoneme but different allophones. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 00:37, 22 May 2008 (UTC)

LOL! Yes I do have it backwards!  :) Ok, I've made a backup of this eternally confusing fact to my page... may I delete this section now? — robbiemuffin ^{page talk} 13:52, 22 May 2008 (UTC)

You mean this section in the talk page? There's no reason to delete it. Other editors may make the same mistake. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 00:27, 23 May 2008 (UTC)