Jump to content

Help talk:IPA/Spanish

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia


voiced /θ/

[edit]

¿why is the voiced realization of /θ/ shown as [θ] instead of [ð]? Brawlio (talk) 20:31, 24 July 2024 (UTC)[reply]

Because ⟨ð⟩ commonly represents the approximant realization of /d/ in transcription of Spanish. Even in narrow transcriptions that mark the voicing of /θ/, the voiced allophone is usually transcribed with ⟨θ̬⟩. But this key no longer marks the voicing of /s, f/ so there's no reason to mark the voicing of /θ/. Nardog (talk) 20:35, 24 July 2024 (UTC)[reply]
not true 75.143.18.234 (talk) 06:19, 8 February 2025 (UTC)[reply]

⟨r⟩ after /θ/

[edit]

i believe the help page says ⟨r⟩ after /θ/ in a word like lazrar would be [ɾ], but i was wondering whether it's actually pronounced as [r] and wasn't mentioned because its occurrence is as uncommon as i think it is. it seems like it could go either way since it would be [r] with seseo or [ɾ] with voiced /θ/ like fricative /d/ Brawlio (talk) 21:26, 24 July 2024 (UTC)[reply]

I think the [r] vs [ɾ] footnote's just slightly off. After a heterosyllabic consonant, only [r] appears. [θ] can't form an onset cluster with [ɾ], so lazrar must have [-θ.r-]. I don't have immediate access to the English version of Hualde's The Sounds of Spanish, but the Spanish version explains this. And /l, n, s/ are just the most common consonants for [r] to appear after.
I'm not sure how I'd rewrite the footnote, but maybe something explaining that only stop phonemes + /f/ can form onset clusters with /l/ or [r] would work; that's how it's explained in the Spanish version of (Hualde 2005) in section 4.3.2. Or you could list out all the consonant phonemes, or even list out the different letters that represent these phonemes as the DPD does.
And yeah, -zr- must be extremely rare word-internally in Spanish, in the DLE lazrar is the only word that has it. Erinius (talk) 00:09, 25 July 2024 (UTC)[reply]
i see, the DPD link you shared provides the example of Azrael pronounced with a [r] Brawlio (talk) 03:31, 3 November 2024 (UTC)[reply]

⟨r⟩ after /x/

[edit]

⟨-jr-⟩, such as in lojro, is another rare word internal sequence that the help page seems to be implying is pronounced /xɾ/ but i'm not sure whether that's accurate Brawlio (talk) 18:01, 27 October 2024 (UTC)[reply]

Based on spelling I'd assume /x.r/, since typically Spanish is described as having only /b d g p t k f/ before the tap in onset clusters - but, as that dictionary page mentions, lojro is related to locro, a loan from Quechua which would have the tap - and in this cooking video the host pronounces it once, with a tap, at like 7:03. Erinius (talk) 05:54, 28 October 2024 (UTC)[reply]
his pronunciation at 6:55 kind of seems like more of a tap, while i think his pronunciation at 7:03 is a little bit more ambiguous. perhaps it's the proximity to the /x/, which i think may be more of a [χ~ʀ̥] realization, which may be expected given the possible quechua influence. but when listening at 0.25 speed, i'm not so sure it's his /ɾ/, but perhaps not quite his /r/ either. definitely need more data Brawlio (talk) 18:16, 28 October 2024 (UTC)[reply]
I think this discussion would be more relevant to Spanish phonology. It won't affect more than a few transcriptions linking to this guide, and there are already thousands of them. Sol505000 (talk) 10:30, 3 November 2024 (UTC)[reply]
I'd be surprised if it affects even a single transcription to be honest. Erinius (talk) 04:04, 4 November 2024 (UTC)[reply]
https://doi.org/10.7557/1.9.1.5321 apparently onset jr- (and zr- in Spain) clusters rate as much more acceptable than sr-. Erinius (talk) 22:06, 16 November 2025 (UTC)[reply]

[ᶞ] (and [ᶿ]) in final position

[edit]

Should we replace final ð, which is highly variable by ᶞ? Arroyo Hernández (2009: 67) distingue entre 6 articulaciones distintas: [d] oclusiva sonora; [ð] fricativa sonora; [ᶞ] fricativa sonora débil; [ᶞ̥] / [ᶿ] fricativa ensordecida débil; [θ] fricativa interdental sorda; y, [Ø] cero fónico. — Jɑuмe (dis-me) 10:33, 29 October 2025 (UTC)[reply]

For instance, ciudad ('city') could be transcribed as [θjuˈðaᶞ]. In my opinion it is better this representation as it is in between its full pronunciation and cero fónico, which is especially common in the -dad suffix. — Jɑuмe (dis-me) 11:01, 29 October 2025 (UTC)[reply]

I think ⟨ᶞ⟩ would be somewhat confusing given the superscript often signifies not optionality but a modification informed by the notational sound to the previous segment. In fact it's the latter usage that was officially part of the IPA at least from 1932 to 1989.
Some article do already use "(ð)" though, like Bellreguard, so it might be a good idea to incorporate it into the key, preferably specifying when it's appropriate after a survey of reliable sources. Nardog (talk) 12:32, 6 November 2025 (UTC)[reply]
The (ð) in Bellreguard is probably because it comes word-finally after an R. Erinius (talk) 14:14, 8 November 2025 (UTC)[reply]
How is that different? Are you saying elision is more common after /r/ than after a vowel? Nardog (talk) 16:01, 8 November 2025 (UTC)[reply]
I'd assume so, since it's the 2nd element in an unusual coda cluster. But you're right, RS's would need to be consulted. Erinius (talk) 00:22, 10 November 2025 (UTC)[reply]
(ð) is perfectly fine IMO, like (n) in Dutch. Or, better yet, just use a plain ð (I'm not using the angbr IPA template on purpose here) without the parentheses as the full pronunciation with [ð] is always correct (and in Spanish any [ð] can be elided when final, which is not at all the case with [n] in Dutch, which is omitted after most instances of /ə/ and, more variably, when the first /n/ in the sequences /n#n/ or /n#m/ (with assimilation of POA) gets degeminated, ⟨#⟩ obviously stands for a morpheme or word boundary here). In addition to that, there could be a degree of regional variation in the broadness of [ð] dropping (though it's always more or less informal, AFAIK) which we can't really account for in this guide. Sol505000 (talk) 15:41, 9 November 2025 (UTC)[reply]
I'm not sure about how informal it is dropping ð, but it is regarded as conversational Spanish and it is almost universal in southern Spain as well as in parts of northern Spain (where it can also be devoiced). In fact, I've heard Queen Letizia, who is from Asturias, dropping it or devoicing it in some of her speeches which means final d is somewhat special in Spanish pronunciation... IMO it could be left without parentheses but a note might be necessary to inform about its high variability. — Jɑuмe (dis-me) 08:53, 11 November 2025 (UTC)[reply]
[ð] devoicing is not restricted to Spain either. I've occasionaly heard it from Mexican actors. As far as Queen Letizia is concerned, it could be just a Madrid accent feature that she picked up. After all, it's just a replacement of one phoneme (/θ/) for another (/d/), which she both has in her phonemic inventory. I'm fine with a note. Sol505000 (talk) 17:13, 11 November 2025 (UTC)[reply]
Devoicing and elision (in this case less than in southern peninsular) is somewhat widespread in northern Spain, from the Basque Country to Cantabria and parts of Asturias. I don't think she got it from living in Madrid, she probably developed it further from there, and Mexico as well (but who knows you could be right). Nowadays you can find all pronunciations in Madrid, so it might originated in another place. I'm not confident to say where devoicing originates in Spanish though, other than North Central Spain (you can check this source Phonetic variation in voiced obstruents in North-Central Peninsular Spanish). Devoicing in Catalan (Northeast) is different since fortition (ciutat) is prevalent and Galician (Northwest) only allows medial final d but not utterance final ð (cidade). Asturian seems to be pro-elision (ciudá). — Jɑuмe (dis-me) 18:43, 11 November 2025 (UTC)[reply]
That's an interesting study. Devoicing of specifically word-final -d has of course been noted in parts of northern Spain and to an extent in Madrid. I found Hualde and Eager 2016 and remembered Molina Martos 2016 which examine this. Devoicing could maybe use a note, Hualde and Eager find it to be pretty common in Valladolid, but Molina Martos finds devoicing to be less common than eliding or retaining [ð] in Salamanca (Madrid).
I'm surprised there isn't already a note at least about final -d elision considering how widespread it is. Erinius (talk) 19:10, 12 November 2025 (UTC)[reply]
Thanks, yes you can add the notes if you can and improve the note about final obstruents it is a bit confusing. — Jɑuмe (dis-me) 09:03, 16 November 2025 (UTC)[reply]

The sequence [θð]

[edit]

The phrase 'Rodríguez de Silva' should, according to this guide, be transcribed [roˈðɾiɣeθ ðe ˈsilβa]. Isn't a full assimilation into a long voiced fricative ([roˈðɾiɣeð ðe ˈsilβa]) more usual in Spain? I know that we use ⟨ð⟩ only for the allophone of /d/, but still. Sol505000 (talk) 23:54, 3 December 2025 (UTC)[reply]

I believe that would be common, at least from what I've heard. I may have seen sourcing but I can't recall any. Also, I noticed that article's transcription has "Silva y" as [ˈsilβaj] - that is of course how it would typically be pronounced, but I don't think this key provides for it. Erinius (talk) 03:49, 4 December 2025 (UTC)[reply]
That's right, it seems this guide (unlike Galician) though does not transcribe voicing anymore. — Jɑuмe (dis-me) 04:48, 4 December 2025 (UTC)[reply]
Yeah, that's kind of what I mean - diphthongizing /ai/ but failing to assimilate /θd/ to a single, long voiced fricative looks a bit strange. At least that's how many people would read that kind of transcription anyway. Sol505000 (talk) 16:12, 10 December 2025 (UTC)[reply]
Retrieved from "https://en.wikipedia.org/w/index.php?title=Help_talk:IPA/Spanish&oldid=1326739850"