Help talk:IPA for Spanish

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Spain (Rated NA-class)
WikiProject icon This page is within the scope of WikiProject Spain, a collaborative effort to improve the coverage of Spain on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
 NA  This page does not require a rating on the project's quality scale.


Orignal Source : es:Transcripción fonética del español con el IPA

Variety of Spanish[edit]

So, do we use the dialect of Spanish used almost exclusively in northern and central Spain, or the one used in every other country that speaks Spanish natively? (regarding, say, seseo and such.) Red Slash 23:36, 28 October 2013 (UTC)

We use Castilian, because, apart from the relatively minor distinction of /b/ vs /v/, you can deduce the pronunciation of any other dialect from it, but you can't do the reverse. Castilian is also the prestige dialect, and the form most familiar in the UK. This is the consensus that was established when we worked out this key. If you want to indicate the local pronunciation of a place name, you can do so by adding 'local' as a parameter to the IPA-es template, but generic Spanish words should be in Castilian. — kwami (talk) 00:04, 30 October 2013 (UTC)
I'm going to ignore the comment about prestige--clearly that has no place whatsoever in a discussion taking place for an encyclopedia. Even if Castilian Spanish is the form most familiar to people in the UK, which is debateable, it's far less influential (of course) on a global scale, and I believe we do write for a global audience. Again, it's a minority dialect. And any dialect of native Spanish speakers keeps /b/ the same as /v/ - I don't know what you're referring to. So those arguments don't do much for me. I can understand your concerns about, if we transcribe "z" as [s] for geographically "neutral" articles, that doesn't let people see that it's pronounced differently in northern Spain. Of course, transcribing the z as a theta makes them think that most Spanish speakers use it as a theta, which only those who bother to scour in the footnotes will likely see.
Anyway, getting down to brass tacks... My question is why you removed--flat-out removed--the [h] sound, which is how almost every person who speaks Spanish pronounces the soft "g" and the "j". Why did you choose to bury that phoneme, which is the supermajority pronunciation, while giving prominence to a minority pronunciation in the table? Another question--the conversion to yeísmo is almost complete, even in Spain, and showing a distinction is really rather unnecessary IMO - why did you decide to phrase it in such a way that leads people to believe that most speakers have not made that transition? (I quote: "For terms that are more relevant to regions that have undergone yeísmo ... ") We're not trying to give an exhaustive list of Spanish phonological concepts and history, much less a crash course in "prestigious" versus "less prestigious" dialects. Why not phrase in a way that clarifies the relative relevance of yeísmo versus the lleístas? My third question would of course be the z/s distinction. Listen, our readers do not need to know the history here regarding seseo; they just see a theta and they think, oh, that must mean "th". But then they will have the wrong idea of how it is pronounced by (the vast majority of) actual Spanish speakers.
In summary, I'm absolutely not suggesting that we remove the minority pronunciations. Not even remotely. But it makes no sense to use a minority standard for our articles and use a standard for this guide that effectively dubs the one used in twenty different countries by 90% of native Spanish speakers as a "regional" or (even worse) "local" dialect, and worst of all, demotes it to a footnote. Latin American Spanish is Spanish. The Spanish of Spain is also Spanish. In no way, shape or form does it make sense to exclude [h] from the table just because "only" 90% of Spanish speakers use that sound. We can talk about some of the other stuff later--I'd love to hear your thoughts on this so far. Red Slash 22:40, 30 October 2013 (UTC)

We're not discussing how to set up the key. The key has already been set up, and has been used as the basis for pronunciation in 3,000 articles. What you are proposing is that we change the key, and thus change those 3,000 articles. If you can convince people to do that, fine, but it's not okay to change the key and leave those articles unsupported. — kwami (talk) 23:39, 30 October 2013 (UTC)

I will set it up so that nothing in the key contradicts anything in any of the articles. Deal? Red Slash 03:10, 4 November 2013 (UTC)

trigo and the voiced velar fricative[edit]

If trigo is really pronounced ANYTHING like "wall", than it is NOT a voiced velar fricative. It is just not. A voiced velar fricative is pronounced roughly like go, but without completely blocking air flow on the g, exactly as I had written. Either the symbol or the description needs to be changed. I tried changing the description, but some obstinate editor changed it back again.

Etoombs (talk) 06:35, 9 December 2013 (UTC)

  • You have my apologies. Red Slash 00:47, 10 December 2013 (UTC)

ll and [ʃ][edit]

Currently footnote 4 says:

In Rioplatense Spanish (roughly southern South America), the ll is typically [ʃ], a sound which has no perfect English equivalent but is close to measure.

This text was introduced last December by Red Slash. The cited article in the Journal of the International Phonetic Association does not even mention [ʃ].

Also, the text is at least a little misleading: The s in measure is a Voiced palato-alveolar sibilant, [ʒ]. The unvoiced version [ʃ] is the common one in Argentina, although I believe [ʒ] is also used. If there is some other subtlety to the Argentine ll, I'd be fine with noting it, but I want to see a citation. —Remember the dot (talk) 22:14, 15 February 2014 (UTC)

Thank you! Okay, that is definitely not cool that I lazily kept the same citation that there was before. I apologize, that's not okay. My understanding of the Argentinian accent from the Argentinians I've known is that "ll" and "y" are both pronounced roughly the same as the French "j" (which again roughly is the same as "measure"). Brazilians I think use "shh" but I'm not sure that Argentinians do (I thought it was the case that the first syllable in "lleno" sounds like the French "J'ai"). I thought it was common knowledge, but I may have mixed up my symbols and in any case if you disagree then I may well be wrong; we'll have to get a source. Red Slash 21:15, 16 February 2014 (UTC)
See Rioplatense Spanish#Phonology where this paper is cited. — Lfdder (talk) 02:15, 17 February 2014 (UTC)
Also see [1]. — Lfdder (talk) 02:22, 17 February 2014 (UTC)
Alright, go for it. Always glad to learn. smile Red Slash 05:27, 17 February 2014 (UTC)

h vs x[edit]

The difference is not phonemic and we should be supposed to follow Castillian norm, except when it's relevant to do otherwise. Why, pray tell, transcribe it as [h] in Latin America? Most Brazilians don't use [ʁ] either, but still in our transcriptions of Portuguese people are happy to share the same symbol with Portugal – and the two variants are much more distant and proud of themselves than Spanish-speaking people tend to be any used to.

Just an opinion: I'm generally against these sensibilities if people know that a certain phonological characteristic merges to a single sound in the ex-colony, or otherwise. Like not transcribing the Portuguese + Rio de Janeiro palatalization in Brazilian Portuguese IPAs, or insisting on have European Portuguese IPAs without the Brazilian palatalization in ti/di. The issue of ce vs. se or ll ≠ y vs ll = y to me is pretty much the same. It just reduces foreigners' exposition to more knowledge in our languages. Uniformity is good and more inviting, not to say how many Latin Americans might have the allophones more typical of Europeans, and might like palatalization, or dislike yeísmo... But I digress. Srtª PiriLimPomPom (talk) 12:27, 9 March 2014 (UTC)

On the other hand [h] could be the general sound in Caribbean spanish, but not in South America and Mexico, where the common sound for literate people is [x] Elchsntre (talk) 12:35, 10 March 2014 (UTC)
Then it's even worse than the Brazilians' and Portuguese people's phobia of the palatalization pattern of each other. *facepalm* They use it in Latin America and widely so, and it's held as standard in important variants. Correcting it, now. Srtª PiriLimPomPom (talk) 03:41, 11 March 2014 (UTC)
Even the Spain-centric RAE admits that in "amplias zonas de Hispanoamérica" the "h" sound dominates. [2] The reason it is explained this way is so that places where the "h" pronunciation predominates, we don't tell our readers how someone from Madrid would pronounce it, but rather how someone from that area would pronounce it. Nobody cares how Juan Carlos I would pronounce Isla de la Juventud, they want to know how someone living there would pronounce it. (This is tricky because when transcribing these sounds, we have to just pick one. It's different than with the seseo, where the deal is that "c" and "z" simply make an "s" sound, no questions asked, and it's absurd to suggest that Venezuela is pronounced with a "th" sound when basically no one living there would ever say it that way. Venezuelans get to decide how their name is pronounced in their own language.) Red Slash 22:30, 12 March 2014 (UTC)
But that's not what we've agreed to do. We give the *Spanish* pronunciation, not the Venezuelan or Cuban. This is because (apart from /b/ vs /v/, which survives is a few places in Spain), local pronunciations are predictable from the Castilian. If we want to give the local pronunciation, we use the "local" parameter of the IPA template, so it doesn't claim to be "Spanish". This is how we handle German, Russian, French, Italian, Japanese, Mandarin, Burmese, Tibetan, and other languages with significant regional differences. Of course, we could agree to change our approach for Spanish, but that requires some thought. Are we going to only pretend to give local pronunciations by choosing [h] over [x], [s] over [θ], and [ʝ] over [ʎ], as if that were all there were to it? What should we call our system then, faux-Spanish IPA? Or are we going to actually work out the phonology of the local dialect of every town and place that we transcribe? Are we going to support Mexico as a legitimate form of Spanish, but not Buenos Aires? What of people, say someone born and raised in Mexico but now living in Buenos Aires? I think we need to agree on that first, and only then change the IPA table to support non-Castilian pronunciations. — kwami (talk) 23:46, 12 March 2014 (UTC)
We are conflating (at least) two issues here.
  1. I have yet to see or hear a single reason why we need to avoid having "h" in the table. This is the IPA for the Spanish language, and a majority of people who speak the Spanish language make the [h] sound when they say "jota". It is unreasonable to pretend the sound does not exist.
  1. And no, local pronunciations are not predictable from the Castilian unless you know that Castilians use [x] where most Spanish speakers use [h]. Or that when Wikipedia tells people Venezuela is pronounced "Beneθuela", it's actually "Benesuela" for 90% of Spanish speakers (including Venezuelans!). (I'm using this as a reducto ad absurdium--obviously the Venezuela article has used the seseo pronunciation for some time.) Remember: the only reason why we would even include the pronunciation guide is so that people know how the word is pronounced. If someone already knows the rules, they won't be confused. If someone does not yet know the rules and they see a θ, they will assume that that's how you pronounce it. They won't know that they can deduce the actual pronunciation from the Castilian version.
I actually don't know why we're bringing up the idea of the θ, since the version that was stable for a couple months very clearly stated that in most of the Spanish-speaking world, "z" and soft "c" make an "s" sound, but in most parts of Spain they instead make a θ sound. We can discuss the wording and all that, and that's very good, but we all agree on that, right? But what is completely illogical is to hide the "h" pronunciation completely under the rug as if it doesn't exist. It does. And it should be listed here.
I would love logical reasons for reverting this edit, if someone's going to do it. There are a fair few things in there, most of them clarifications for people who come here and need help understanding Spanish pronunciations. (Like making it explicit that [x] has no English equivalent. "Loch is not at all an English word; it's Scottish (Gaelic), and it is a loanword in English that only gets that [x] sound if pronounced by someone who is trying to emulate the Scottish pronunciation. Just like "arroz con pollo" is pronounced with a trilled "rr" only if it's said by someone who knows Spanish. So let's make it obvious, even though it's hidden in a reference, that in the column labeled "English approximation" that there is no English approximation.) Red Slash 01:57, 13 March 2014 (UTC)
If you want some invented melange of dialects, then instead of "Spanish pronunciation", we should say "random collection of Spanish sounds". And of course you have to be able to predict the changes to be able to predict the changes: that's what "predict" means. But with rare exception, we do not give the local pronunciation. Rather, we give the Castilian, with efforts by some people to accommodate the three differences that they know of as foreigners. That's not even close to giving the actual pronunciations of these towns. If we add [h] for "j", then why not add [ʃ] and [ʒ] for "y"? Why not give "pan" as an example of a word with [ŋ] in it? Why not give "piel" as an example of a word with [ɾ] in it? Why not "perro" for a word with [x] in it? We don't accommodate Spanish dialects, we just pretend to, which is dishonest. When we set up this table, we had consensus to use Castilian pronunciation. If we want to change that, fine, but first we should come to consensus to do that, and then we should use real Spanish. — kwami (talk) 02:32, 13 March 2014 (UTC)
Not to say how what he is trying to do is even patently wrong in the data he's adding. /ʃ/ is indeed an English phoneme, and the [ʒ] in measure is nothing like the Tupí-Guaraní(?)/Argentine(?)/Brazilian/Catalan [ɕ] that is an unrounded, further palatalized [ʃ], not a demi-voiced one! He might mean the Tupí-Guaraní(?)/Argentine(?)/Brazilian/Catalan [ʑ], but that is another matter entirely and another phone system, not the one used in Buenos Aires. If he's so sure of pushing some sort of Wikipedian linguistic neobolivarianism (just kidding), he should research a bit more before on our American varieties.
Further, he's trying to tell English speakers to do all we DON'T want them to in regards to the vowels, that is diphthongize the /e/ and /o/ in every instance they see it. Because Americentrism is good, when he could propose to just add "*linky* Received Pronunciation *linky* | RP" to the "get", but seemingly it doesn't matter if there's none of that in America. OR he didn't research the English phonology articles to be sure of what he is talking about, what is important because our own native language and dialect creates bias when we study such things. Srtª PiriLimPomPom (talk) 12:20, 13 March 2014 (UTC)
Also, what's with 'dis' for Spanish /ð/, later changed to 'No standard English equivalent'? — Lfdder (talk) 13:49, 13 March 2014 (UTC)

/z/ ???[edit]

In my linguistics courses, I was taught that the /z/ or the voiced alveolar sibilant, does not exist in properly spoken Spanish. This should be at best moved to "marginal sounds," or marked as dialectical. What say you? I need a source before I make the edit. Help? Iamvered (talk) 18:20, 20 March 2014 (UTC)

[z] is not marginal; it's allophonic. It occurs often. — Lfdder (talk) 18:43, 20 March 2014 (UTC)

I don't have a citation for this, but.... The sound used for "s" in "prison" is, in some western and southern U.S. dialects, in between the sound used for z in "zombie" and the sound used for s in "salamander". Does the IPA have a symbol for the hard Z sound, or is that what this is supposed to be? If it is, then I propose changing the example to something else, like "zombie", or if we can't use an example with a "z" in it, "xylophone".