Talk:Spanish phonology/Archive 3

Archive 1

Archive 2

Archive 3

Article Overly Long and Lengthy

Much the allophony is needles in any phonology article. that's a given in any language. Otherwise, this 'd be titled "spanish phonetics". Yoandri Dominguez Garcia 11:13, 21 February 2019 (UTC)

It would be a bit strange to have a separate article on a language's phonetics, given the close relationship that phonetics has with phonology. Both are in the scope of the articles titled "XX phonology." If there were a better name given this scope, we'd use it. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 17:30, 21 February 2019 (UTC)

When we say "the phonology of language ____", "phonology" there means not the distribution and variation of phonemes as juxtaposed to phonetics but the sound system of the language as a whole, including both phonetics and phonemics. See e.g. [1]. Nardog (talk) 04:55, 22 February 2019 (UTC)

Can we have a dedicated Wikipedia article on the coda simplification in Spanish spoken around the world?

I noticed that syllable- or word-final dropping of /s/ in Spanish-speaking Latin America, Iberian Peninsula, and Africa (Canary Islands) are prevalent, along with YouTube clips on so-called "how to speak Spanish like a native". And there is also word-final dropping of /ɾ/ in some speeches. These days, I noticed that this information is well-expressed in many Wikipedia articles pertaining to Spanish dialects. I think all of theses should be addressed in a new separate article. I'm not a Spanish speaker myself, so I'm requesting this. Thank you. --Komitsuki (talk) 10:40, 8 May 2019 (UTC)

I don't know if that would warrant a whole article. Perhaps you could draft something in sandbox and we can go from there. If we decide not to do a full article, content that you draft might still be relevant here. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 19:44, 8 May 2019 (UTC)

Insertion of velar stop between /s/ and /r/

I've noticed that when /s/ is followed by /r/, speakers insert a velar stop in between the two sounds. For example <tienes razón> /ˌtje.nes.raˈson/ is pronounced [ˌtje̞.nes.g̥raˈson] and <ruidos raros> /ˌrwi.ðosˈra.ɾos/ is pronounced [ˌrwi.ðosˈkra.ɾos]. I am, by no means, an expert Spanish and linguistics, so I was wondering if someone could explain this to me, and if I'm not just hearing things, then I could add it to the article. User.name.here (talk) 21:58, 26 June 2019 (UTC)

I suspect that a source that discusses this phenomenon might have an explanation as well. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 22:36, 26 June 2019 (UTC)

That must be peculiar to a regional dialect. --Jotamar (talk) 17:59, 27 June 2019 (UTC)

I'm not a fluent speaker of Spanish, but I am learning. I don't live around many native speakers. I do, however, listen to Spanish music. I have heard this phenomenon in songs by Ricky Martin (a Puerto Rican) and Maluma (a Colombian Paisano), which is why I don't think it's dialectal, but like I said before, I'm not an expert, so I could be wrong. User.name.here (talk) 23:16, 27 June 2019 (UTC)

@User.name.here: Aren't you just mishearing s-aspiration? A speaker that exhibits it would say [ˌtjeneɦ‿raˈson] ([ˌtjeneɦ‿raˈθon] in Spain) and [ˌrwiðoɦˈraɾoh] (or, perhaps, [ˌrwiðoɦˈraɾos] since this is variable). Spanish doesn't have phonemic glottal consonants, but /x/ is glottal for millions of speakers and there may be allophonic overlap between it and the aspirated allophones of /s/ in the syllable coda. There's additional complication in that aspirated /s/ can be actually velar [x] before /k/.

I've never heard of the phenomenon you're describing. Kbb2 (ex. Mr KEBAB) (talk) 15:09, 1 July 2019 (UTC)

@Kbb2: No, I'm sure it's not s-aspiration. Thanks for your help though! Also, here's a link to a song where I'm hearing the phenomenon I described: https://youtube.com/watch?v=iOe6dI2JhgU. I'm hearing (or mishearing) the phenomenon at approximately 1:18 in the video. User.name.here (talk) 19:50, 1 July 2019 (UTC)

@User.name.here: It's a normal [z], perhaps with reduced friction. It sounds like a cross between English [z] and the fricative allophone of /ð/ (as in within). I'd transcribe tienes razón as pronounced (sung) by Martin as [ˌtjenez‿raˈson]. I don't hear any stops before /r/.

I've heard this used in parts of Argentina for /s/ in syllable coda. Instead of being aspirated, the friction is reduced and they sound somewhat lisped (but in a different manner than the English or European Spanish /θ/). In Argentina, this allophone is in a free variation with the aspirated [h] (and [ɦ] before voiced consonants, such as /r/ or /n/).

There's another allophone that can be used (e.g. in Mexico), which is a kind of a voiced retroflex fricative [ʐ]. It's used specifically before the trilled /r/, and can be thought of as a fricative onset of the trill. Speakers with this allophone would pronounce tienes razón as [ˌtjeneʐ‿raˈson]. Both count as allophones of /s/ and may be used in other countries, which wouldn't be that strange. In some other areas, the retroflex fricative itself is the realization of /r/. Kbb2 (ex. Mr KEBAB) (talk) 08:11, 2 July 2019 (UTC)

I assume what User.name.here is hearing as [k] is a devoiced part of the beginning of [r]. I perceive both the end of /s/ and the beginning of /r/ to be devoiced, with the result that the first contact of the trill sounds like a voiceless plosive. Nardog (talk) 07:57, 3 July 2019 (UTC)

@Nardog: Then what is an underlying sequence of /s/ and /r/ may be actually [zz̥t̞r̥r] (with the extra-short diacritic implied on every single symbol) in Martin's pronunciation. There may be a slight epenthetic stop between /s/ and /r/, but it's alveolar. Kbb2 (ex. Mr KEBAB) (talk) 22:01, 3 July 2019 (UTC)

Thank you all for your help, I understand the pronunciation now! User.name.here (talk) 04:05, 18 July 2019 (UTC)

Possible allophony of [β] and [w]?

The voiced bilabial approximant and the voiced labio-velar approximant are certainly very similar sounds. They could easily merge and they are in fact allophones of each other in some languages, e.g. Flemish Dutch. Regarding Spanish, I also think that the functional load of the distinction is very low. The only minimal pairs I can think of right now would be words with initial hu- + vowel, but these are pronounced /ɡw-/ in some accents. So at least for these latter accents, is the distinction relevant at all? (I'm not an expert on Spanish, I'm just wondering.) 90.186.72.23 (talk) 14:24, 10 January 2020 (UTC)

AFAIK, there's not much risk of the two merging in Spanish. You might be interested in Ohala & Lorentz's "The Story of [w]: An Exercise in the Phonetic Explanation for Sound Patterns" (1977), cited in the article. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 15:09, 10 January 2020 (UTC)

I've never heard [w] in that context. It has a velar counterpart not present in the plosive allophone of /b/ and, AFAICS, requires a more precise articulation than the bilabial approximant (compare [xaˈβon] and *[xaˈwon] for jabón). In careful pronunciation this /b/ can be turned into a plosive: [xaˈbon]. Remember that the difference between [xaˈβon] and *[xaˈwon] is nothing more than velarization. Kbb2 (ex. Mr KEBAB) (talk) 20:05, 18 January 2020 (UTC)

la madre [la ˈmaðɾe] ('the mother') vs. las madres [læ̞ː ˈmæ̞ːðɾɛː] ('the mothers')

Why would the first vowel in "madre(s)" would change in plural if there is no change in its surroundings? It doesn't make sense.

--200.57.197.151 (talk) 16:08, 22 January 2020 (UTC)

It assimilates to the frontness of the first vowel (apparently, /a/ becomes more front instead of more open, which it can't become as it's already an open vowel). Kbb2 (ex. Mr KEBAB) (talk) 06:13, 23 January 2020 (UTC)

It is a type of vowel harmony. --Jotamar (talk) 00:25, 25 January 2020 (UTC)

[s̄]

In section "Realization of /s/" the notation [s̄] is used several times to represent a voiceless apico-dental grooved frivative. Though ⟨s̄⟩ is characterized as an "ad hoc symbol" I believe this is an unfortunate choice because in IPA, the macron above a segmental symbol is used for a different purpose: It denotes a mid level tone. (This is not the place to discuss whether voiceless sounds can carry tone.) As there is no single IPA diacritic to turn a dental slit fricative [θ] into a grooved one, the notation [s̺̪] or simply [s̪] with the diacritics for (apical and) dental seem to be the appropriate IPA notations. Both ⟨s̺̪⟩ and ⟨s̪⟩ display fine in my browser, but this is certainly not a requirement as long as the symbols are the correct ones. — Also note that slit [θ] and grooved [s̺̪]/[s̪] don't seem to contrast in any Spanish dialect ("It occurs only in dialects with ceceo."), and that the IPA symbol ⟨θ⟩ doesn't necessarily denote a slit fricative, but is defined as any pulmonic dental fricative, so we might as well use ⟨θ⟩ throughout and explain the occurrence of two phones in words. Love —LiliCharlie (talk) 09:33, 6 May 2020 (UTC)

I like the idea of using ⟨θ⟩. Outside of whatever sourcing that parses the phonetic distinction between [θ] and Spanish's voiceless apico-dental grooved frivative, the overwhelming tendency is to use ⟨θ⟩ for the sound of Spanish. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 14:42, 6 May 2020 (UTC)

Phonemic transcription

I'm not saying that my (reverted) version of the phonemic transcription in the text is the only possible one. But user:Nardog's one is untenable. Syllabification, a key phonological feature, is shattered with things such as fuˈeɾte. We could write /'fueɾ.te/ or perhaps /'fweɾ.te/, but certailnly not /fuˈeɾte/. Even worse is the fact that apparently the aforementioned user thinks that disyllabic words always have phonological stress, and that monosyllabic words never have it. This simply reflects ignorance about the Spanish language. Lexical stress and syntactic stress diverge only in set phrases. I'm not going to revert for the moment, but the section as it stands now is simply wrong. --Jotamar (talk) 14:31, 6 May 2020 (UTC)

If I'm to understand correctly, you object to e.g. viento being transcribed phonemically as /biˈento/ because phonetically that word is two syllables ([ˈβjento]) and the phonemic transcription makes it look like three. However, if we were to take your suggestion, this would be transcribed phonemically as /ˈbiento/, which seems also untenable to me. Take a look at discutían just a few words later; we transcribe this as /diskuˈtian/, but this phonetically is [diskuˈtian]. Transcribing e.g. viento as /ˈbjento/ would be a problem because the common analysis of Spanish (as I understand it) is that this semivowel is the result of a phonological process that de-syllabifies vowels, not because of an underlying semi-vowel segment.

It would help if we could identify what solutions sources tend to do in this situation.

This whole section is actually unsourced. While we have a nice audio file to go with this transcription, the transcription itself is OR. We do have a phonetic transcription of the exact same passage from Martínez-Celdrán et al (2003) for Castilian Spanish, but no audio file. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 14:55, 6 May 2020 (UTC)

"It would help if we could identify what solutions sources tend to do in this situation": As I explained in the edit summary, most works about the phonology of Spanish use a stress mark, even if they cling to IPA for the rest of the transcription. So, /biénto/, or in any case /'bien.to/ with syllable boundaries rather than just /'biento/. /bi'ento/ is a real eyesore for a speaker of Spanish. --Jotamar (talk) 21:16, 6 May 2020 (UTC)

The audio by Martínez-Celdrán et al (2003) is available here. And the passage is not identical to what we have. Nardog (talk) 21:49, 6 May 2020 (UTC)

I've restored the syllabification and hiatus marking. The stress mark need not necessarily signify a syllable division simultaneously—see e.g. Deutsches Aussprachewörterbuch, which puts it before the stressed vowel even if there's an onset—but since hiatus is obligatory in some (usually morphologically complex) words like dueto and huimos, preserving the syllabification in phonemic transcription makes sense. Of the JIPA illustrations of Spanish varieties, only Coloma (2018) has a phonemic transcription, and he writes ⟨ˈbiento⟩ (he even transcribes discutían and sería without indicating hiatus, probably because there's a morphological boundary, but nonetheless it doesn't seem like a good call).

What do you mean by "Lexical stress and syntactic stress diverge only in set phrases"? Don't mientras, como, entre, etc. bear stress when said in isolation? Nardog (talk) 21:49, 6 May 2020 (UTC)

In isolation? Those words are only used in isolation for metalinguistic uses, the same as el or con; well, mientras is a partial exception, when it means mientras tanto (meanwhile) it is stressed, but that's not what happens in our text. Obviously I'm speaking about the preposition entre and the conjunction como, as used in the text, not about the unrelated homophones, forms of the verbs entrar and comer. And of course sol (as all nouns), dos (as all numerals) and más are stressed words. --Jotamar (talk) 22:55, 6 May 2020 (UTC)

Well, I'm not entirely convinced (I don't think "monosyllabic words never have [lexical stress]", I'm just skeptical of the idea that the absence of stress in function words is a product of the phonemic composition of the words per se and not a product of intonation—unless of course you can show me a source that confirms it), but I've restored your stress placement. Nardog (talk) 23:16, 6 May 2020 (UTC)

English has disyllabic prepositions too, namely: over, under. I clearly perceive those prepositions as unstressed. Don't you agree? --Jotamar (talk) 16:58, 7 May 2020 (UTC)

You seem to be talking about stressed words in sentences instead of stressed syllables within words. Don't we mark word(-internal) stress rather than higher-level stress in foot groups and intonation groups? Love —LiliCharlie (talk) 17:27, 7 May 2020 (UTC)

It's been a while since I read it, but Lavoie (2002) talks about stress distinctions between content and function words. Might be a good place to look. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 17:40, 7 May 2020 (UTC)

We are discussing a phonemic transcription rather than a more loosely defined broad transcription. In other words, all allophony is ruled out, and we indicate "stress phonemes", and don't care about their allophones. I believe that every Spanish polysyllable has at least one syllable that is stressed (carries primary lexical stress), though that syllable may be realized weak due to sentence stress; however, as soon as sentence stress shifts to that word (e.g., for contrast) the stressed syllable clearly shows.

Thanks for pointing me to Lavoie (2002). I think I would agree to not indicating lexical stress of function words in a merely broad transcription. Love —LiliCharlie (talk) 00:32, 8 May 2020 (UTC)

"... though that syllable may be realized weak due to sentence stress": I see it differently, both for Spanish and English, and I guess for any language with lexical stress. Some words are intrinsically unstressed, and some words can be stressed or unstressed depending on their function. For instance, the is always unstressed while this is always stressed; compare the intonation pattern of the book and this book; that is stressed as a demonstrative but unstressed as a relative, etc. --Jotamar (talk) 06:33, 8 May 2020 (UTC)

Some words are intrinsically unstressed, and some words can be stressed or unstressed depending on their function. [...] intonation pattern [...] as a relative, etc. My understanding is that the topic of this discussion is whether certain syllables are stressed (word/​lexical stress), not whether certain words are (i.e. sentence/​prosodic stress). Or are you suggesting we should mark sentence stress, and maybe intonemes, in our phonemic transcription? And do so on the same level and in the same way as we mark lexical stress? — Remember this is supposed to be an emic transcription. Love —LiliCharlie (talk) 10:53, 8 May 2020 (UTC)

Ok, I'm probably not getting what you're saying. Of course I'm speaking of something that is 100% emic. My point is simply that a phonemic transcription like /'la 'kasa/ is wrong for Spanish (do you agree with that?), and for the exact same reason for instance /'kontɾa 'la 'xente/ or /'kontɾa la 'xente/ are wrong too. --Jotamar (talk) 18:33, 8 May 2020 (UTC)

My point is that word stress and sentence stress are two different things that follow different rules. I think it should be /ˈkontɾa la ˈxente/ if only word stress is marked, and if you also want to mark sentence stress you need to indroduce some other notational device, for example /ˈkontɾa la ˈxente/ (as opposed to the emphatic /ˈkontɾa la ˈxente/). Love —LiliCharlie (talk) 18:49, 8 May 2020 (UTC)

/ts/ for ch

Where in Spain do they pronounce ‹ch› like /ts/? I've heard this several times, but I have never found mention of it anywhere. I actually quite like it because it makes the system more symmetric :) For examples, listen here (e.g. "escuché" at ca. 10 seconds, "luchan" at the very end). 178.7.217.119 (talk) 14:27, 20 January 2022 (UTC)

At least in/around Madrid. Erinius (talk) 15:22, 20 January 2022 (UTC)

[ð̞]

How does someone pronounce this correctly? Is it like D and R (English) pronounced at the same time? I hear through Wiktionary sound files that it's something like between a Z, R, and D. 61.247.7.157 (talk) 13:46, 1 February 2022 (UTC)

I don't find it useful to describe one sound as a combination of two or more sounds. In English, the sound nearest to [ð̞] is the "th" of "this, that, these, those". For more precision: give the tongue a "lighter touch" than in English. Kotabatubara (talk) 18:38, 1 February 2022 (UTC)

Extremely marginal exceptions to word-final -r corresponding to a tap in related words

I added the text "Morphologically, a word-final rhotic always corresponds to the tapped [ɾ] in related words" to the section about rhotics on this article. This is accurate for the entire lexicon of almost the entirety of the Spanish-speaking world, however, there is one present-day exception, and there may have been some historical exceptions which as far as I know are unattested.

The present-day exception consists of demonyms used in the Roncal Valley, an area with a population of little more than 1000 people where a Basque dialect used to be spoken. These demonyms take the form -ar in the masculine singular, -arra/-arres in the feminine and plural.

The historical exceptions would've been some words ending in -rre, where the final -e was supposedly dropped and later restored. My source for this is Hualde's "Quasi-Phonemic Contrasts in Spanish" which quotes Penny's "A History of the Spanish Language" as saying an alternation "probably once existed between singular tor and plural torres".

I feel like these exceptions are so marginal that, if mentioned on some general article about the Spanish language, they should be mentioned one about its history or something, and not this page or even the one about dialectal differences, but I'm looking for feedback. Erinius (talk) 05:22, 21 July 2022 (UTC)

How can I put this politely? It's a fascinating fact, and might show up in a comprehensive treatise—but does it really belong in any encyclopedia article? Wikipedia should have (though I haven't seen it) an explicit policy on just how marginal a piece of data can be and still be appropriate to include in the type of introductory article an encyclopedia provides. Kotabatubara (talk) 14:57, 21 July 2022 (UTC)

IPA vowels used here vs. those in the Spanish article

Not using the most precise symbols is misleading. Right now I am teaching a class and this article, as it is now, cannot be used as my students get confused and I had to tell them that the "correct" vowel table is in the Spanish article. That solved their confusion.

In the Spanish Wikipedia article they use ä, e̞ and o̞. But here, my students read a, e and o, which ARE NOT the same sounds. Despite the fact the correct wikilinks are being used in this article for all the vowels, the symbols used are not the required ones and, unless they follow the links, they won't easily realize the difference. George Rodney Maruri Game (talk) 23:42, 24 July 2022 (UTC)

There are some compromises made in our transcription for readability and accessibility. If you would like to revisit the question of transcription conventions, I suggest that you find something more compelling than that you are personally not capable of instructing your students in the pronunciation of Spanish vowels in a manner that allows them to use this article. Personally, I don't find that line of reasoning very compelling. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 22:51, 29 July 2022 (UTC)

Narrow transcription does not necessarily lead to accurate transcription. Humans aren't consistent, and if you actually measure the formants, the "correct" symbol for an instance of e.g. /a/ could be any of [ä, a, ɑ, ɐ, æ], if not more. That level of detail is impossible to maintain and ultimately irrelevant except where the specific quality is the focus of discussion. Nardog (talk) 23:27, 29 July 2022 (UTC)

Rising diphthongs? Really?

I find the pronunciation of the so-called "rising" diphthongs described everywhere clearly wrong. It does not correspond to the actual pronunciation - am I the only one to notice this?

"ie" does not sound like [je] (which is the same, only differently rendered, as [i̯e]), but in my opinion like [ie̯]. Just compare the undeniably different pronunciation of Spanish "pie" (foot) versus French "pied"! In French the diphthong is clearly rising, the "i" clearly unsyllabic and hardly perceptible as an "i"-**vowel** anymore - it is different in Spanish, where it seems to be even more accented than the "e".

Also, please listen when a Spaniard is about to utter "Shit!", it sounds like "miiier" and not like "mjeeer". You hear something similar with "miércoles", "tiempo", etc.

The same is true for "ue". It never ever sounds like a [u̯e], but instead like a [ue̯].

My position is also underlined and confirmed by "hasta luego", where the "luego" comes across more like a ['lʊə̯ɣo], at least when (as is so often the case) spoken very quickly!

And besides:

"Luis": That sounds like [lujs], and "muy" like [muj].

Here, too, the "u" is undoubtedly more prominent - so much for "rising diphthong"!

At the "forvo" website there are many spoken examples; very few of the speakers here offer evidence for the claimed rising diphthongs, quite the contrary.

MWV, Saarbrücken

94.219.186.110 (talk) 00:55, 3 January 2023 (UTC)

Two people can listen to the same voice clips and hear quite different things. While some people on forvo seem to say "pie" as with a syllabic "i", in no case have I heard the "i" sound more accented than the "e". In my own experience with Spanish "in the wild" I've only ever heard "ie" as [je] - but in any case, Wikipedia content should be based on what reliable sources say and not editors' personal opinions and judgments, and reliable sources describe these sequences as rising diphthongs.

I've heard the e's quality in "ue" get changed quite a bit, especially in words like "luego" and "nuevo", and I know there's some phonological literature covering this. That said, the default pronunciation seems to be [u̯e], and that's what the IPA's Illustration of Castilian Spanish describes.

"muy" usually is transcribed as [muj] anyway, and there's no contrast between [uj] and [wi] in Spanish so it makes sense that "Luis" could be pronounced with [uj] as well. Erinius (talk) 06:22, 3 January 2023 (UTC)

"deshuesar"

I reverted [des.w̝eˈsaɾ] back to [dez.w̝eˈsaɾ]. The point of citing deshuesar (as a minimal pair with desuello) is that the [w] of hueso can alternate with [ɣʷ], a voiced consonant, and that /s/ before a voiced consonant is syllablle-final and realized as a voiced [z]. Kotabatubara (talk) 15:36, 12 July 2023 (UTC)

It's not just that it can alternate with [ɣʷ], I think that the one without [ɣ] is nothing but a spelling pronunciation, another one of so many in Spanish. --Jotamar (talk) 21:38, 12 July 2023 (UTC)

Why is Castilian Spanish the default on this page?

There are a lot more Spanish speakers in Latin America than in Spain; I believe Mexico is the most populous Spanish-speaking country. 167.206.19.130 (talk) 14:08, 4 September 2023 (UTC)

I don't edit this page often, and I don't know what is the reason or excuse for that, but you should be aware that Castilian Spanish really stands for spelling pronunciation as typical of Madrid, and the fact is that spelling pronunciations are virtually universal among Spanish speakers in formal registers, so it's not really such a lopsided guideline as it might seem at first sight. --Jotamar (talk) 22:02, 5 September 2023 (UTC)

I used WikiBlame to investigate and found when it was added: First, in this edit on 14 May 2009, @Kwertii: added a "Castilian = default" disclaimer here, then later that day @Aeusoes1: rephrased it to saying "Standard Spanish", based on Castilian, is the default, you added a Template:fact tag, and Ausoes1 then rephrased things again, giving something nearly identical to the current "Castilian = default" statement.

I'm not sure there's any particular reason that statement hasn't been deleted or substantially changed, but I'm not sure it should be either. Erinius (talk) 00:00, 6 September 2023 (UTC)