Talk:Catalan phonology

Untitled

Regarding the ela geminada, is it ever written with the Unicode character Latin Capital Letter L with Middle Dot? [1] Or do most people type the dot independently? A-giau 01:30, 25 Sep 2004 (UTC)

May I suggest a few tables? They make it a lot easier to get a general picture of the sounds. Template:prettytable is very useful for this purpose. Check out Swedish phonology or Dutch language. Peter Isotalo 22:37, Mar 22, 2005 (UTC)

Copyediting guidelines

I'm in the process of formatting the article. This means:

• IPA conversion: almost done, spotted a few mistakes of my own. See IPA in Unicode. If you add to the article, use {{IPA| (whatever) }} to insert phonemic or phonetic notation, even if it is not a specific IPA character.
• Citation forms: for Catalan words (or English words) used as examples, Wikipedia italics; for glosses, "double quotes". For example: llop "wolf". No parentheses or commas.
• For spellings: italics. I'll be changing single quotes (like 'qu') to italics (qu).
• Latin words: I don't think it's OK to cite Latin in Small Caps just because it's traditional to do so in some grammar books, or as if Latin were more important than other languages so as to deserve a different notation.

Input on this would be appreciated. --Pablo D. Flores 13:09, 29 Apr 2005 (UTC)

Please convert the article to being just a phonology. Having a sound-to-orthography table at the bottom of the article isn't a problem, but the focus is way too much on spelling rather than phonology right now. The article is a pretty hard read, too, since it consists of almost nothing but bullet points. Try to convert it into more free-flowing prose. Peter ^Isotalo 08:22, 6 January 2006 (UTC)

Question

Under vowels, shouldn't it say "/ə/ e, è. Only in unstressed position"? I'm not exactly a linguistics expert so I might be wrong, but right now it contradicts the info under the other vowel categories... ugen64 01:05, 10 October 2006 (UTC)

Use of tilde

Why is the Catalan language's use of the Latin alphabet not discussed here? I wanted to find information about Catalan's use of the tilde but can find no such description here, nor in the Catalan language article. This seems a serious omission. Badagnani 10:16, 22 February 2007 (UTC)

Already added in Catalan orthography article. Perique des Palottes (talk) 15:44, 16 June 2010 (UTC)

Lead paragraph

This article needs a lead paragraph. Badagnani 23:24, 17 March 2007 (UTC)

Scope

The article's scope is a bit muddled between phonology and orthography. Other than Latin, all the other phonology pages are just about the phonology. I think there should be a Catalan phonology page and Catalan orthography page, the latter of which would be one and the same with Catalan alphabet. If nobody opposes, I'll probably start it myself in a few days. Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 00:44, 23 July 2007 (UTC)

Vowel harmony

Vowel harmony [2] lists "dialects of Catalan language", but the Catalan phonology does not mention vowel harmony. Is one or the other incorrect? Barefact (talk) 15:57, 26 April 2008 (UTC)

I've never heard of it. The vowel harmony article doesn't source it so I'd say we remove it until someone finds a source. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 18:37, 26 April 2008 (UTC)

Would you consider questioning the fact instead of outright removal? I wonder if there is some meat in it. Barefact (talk) 02:39, 28 April 2008 (UTC)

Maybe you're right... Let's see, a cursory look at Google scholar shows some interesting citations

• Jiménez, Jesús (1998). "Valencian Vowel Harmony". Rivista di Linguistica. 10: 137–161.

There's also this book which has a chapter on vowel Harmony in Romance languages and goes into detail about Valancian VH. The Catalan Wikipedia does have this ma[ that shows vowel harmony in Valencian but the cited source is a geocities website. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 03:11, 28 April 2008 (UTC)

Catalan does not contain vowel harmony. Vowel harmony is when all of the vowels within a word change to become all front vowels, all back vowels, or all round vowels based on a trigger vowel. What Catalan has is vowel reduction, similar to English. Vowels in unstressed positions change to become more unmarked vowels, in this case for Catalan, two of the unmarked vowels are [u] and schwa. Vaaht (talk) 07:59, 19 August 2008 (UTC)

That is one form of vowel harmony, though as the vowel harmony lead section states, vowel harmony is any assimilatory process of a language involving vowels not sequentially connected. This means that any feature of a vowel could transfer to other vowels in a word.

I recommend you check out this book which has a chapter on vowel harmony in Valencian (which hasn't the vowel reduction process you describe) and Andalusian Spanish. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]}`

First off, vowel harmony is an assimilatory process in a language from one vowel to the next over the domain of application (which may be language-specific, but is usually the word) possibly through intervening vowels (so-called "transparent" vowels). It may also be blocked by intervening vowels (so-called "opaque" vowels). One vowel transferring a feature to another vowel non-iteratively is probably more appropriately called "assimilation", not "harmony".

Second, the wikipedia page you linked on vowel harmony is misconstruing what vowel harmony is. Yes, vowels are not always adjacent to each other since there is usually an intervening consonant between them. However, the emphasis made on the fact that harmony is between "non-adjacent sounds/segments" is blatantly wrong when you consider that Autosegmental Theory states that different segments may appear on different phonological tiers. Two of the most recognized tiers are the consonant tier and the vowel tier. These were proposed for two reasons: (1) vowel harmony seems to treat consonants transparently, and (2) languages like Arabic contain verb roots which only refer to the consonants (verb conjugation is typically done by changing the various vowels that intervene between the consonants). If Autosegmental Theory is correct, then two vowels are adjacent on the vowel tier, regardless of whether a consonant intervenes between them or not.

Now to look at the article you linked. If you'll notice, Lloret's examples for the Catalan and Andalusian Spanish both only contain one or two syllables. Since this wiki page isn't concerned with Andalusian Spanish, I will disregard it in my explanation (plus further look at this article by Lloret and Jiménez, seems to indicate that Andalusian Spanish has some type of weak vowel harmony, although it's not completely regular).

What Lloret describes can be reanalyzed as plain and simple total assimilation. The domain of application is only the foot. True vowel harmony extends throughout the entire word, so her three-syllable examples should have harmonized. Note that the [-ATR] harmony examples for the three-syllable words in (20) are ungrammatical. You might say, "Well, maybe they just can't cross intervening [+ATR] vowels." Well, yes, that might be true and is also pretty common. However, it doesn't negate the fact that the only application of "vowel harmony" for her Catalan examples is from one syllable to the next syllable, with no iterativity. Now, if she had given a three- or four-syllable example showing the same process, there would be nothing to discuss.

At this point you might say, "Well, what about the examples in (21)?" Yes, these examples do seem like real vowel harmony. However, Lloret's examples do not demonstrate a corresponding word without the final /a/ that does not contain "harmony" within the same dialect, which makes it conceivable that the underlying representation for the dialect in question does not in fact contain /a/ in these words, but the [-ATR] counterparts. It's possible that what she is describing is a diachronic phonological account of the dialect in question, but she does not present it as such. However, even if it is a diachronic account, this does not explain why the only vowel that alternates is the /a/. The one exception is the vowel /o/ in the word "towel"; however, since she presents no other examples, it is possible that this is the only word showing true vowel harmony. This example is further discounted if we consider that this word is probably a borrowing from another language (Latin maybe?), and loan-words are frequently known to interact uniquely with regards to phonological processes.

Thus, given all of this evidence, I must conclude that this dialect of Catalan does not contain "vowel harmony" as such. It may contain total assimilation of the /a/ to a preceding [-ATR] vowel; however, the evidence for that is also weak given the lack of morphological alternations demonstrating this assimilation. Perhaps examples from adjectives in this dialect would back up Lloret's analysis more than the data that she currently gives.

For further reading on true vowel harmony, I recommend that you read Bakovic's dissertation on vowel harmony, demonstrating regular vowel harmony in multiple languages, as well as vowel transparency, opacity, and other related processes. Vaaht (talk) 10:19, 19 August 2008 (UTC)

You seem to be working backwards from a preexisting conclusion. You've decided that there is no vowel harmony in Catalan and your refutations of the article amount to "I don't buy it, show me more." At the very best, you've shown that the existence of vowel harmony is debatable, not that it's definitely absent. If you've got the research capabilities and the know-how, perhaps you could find a review or discussion of Lloret's work. I'll check out the dissertation when I get a chance later today. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 15:51, 19 August 2008 (UTC)

I don't have access to the dialect in question, otherwise, yes, I could conduct my own research to verify this claim (I'm currently studying Linguistics in grad school and so I am familiar with linguistic academic research, particularly field work). However, it might be possible to talk to either Lloret or Jiménez via e-mail to verify if they have checked adjectives when they were doing research on this topic. I'll let you know if I have any luck with that. Vaaht (talk) 01:12, 20 August 2008 (UTC)

IPA Transcriptions

This page concerns me because a lot of the IPA transcriptions as well as the IPA consonant chart are blatantly wrong. There is no symbol corresponding to the voiceless alveolopalatal fricative in words like 'baix' [baʃ] and 'caixa' [kaʃə], yet these are very common pronunciations in Barcelona Catalan. I believe the palatal fricatives and affricates are not found in the majority of the dialects (although feel free to correct me; I only have experience with Barcelona Catalan and Valencian). Also, the transcription for 'poble' should be [pɔpːlə] (Barcelona dialect, and possibly others). The source of my information is first-hand experience staying in Barcelona living with a Catalan speaker from just outside of Barcelona as well as a Valencian speaker, as well as the following article: "Els sons del català" by Joan Julià i Muné out of the book "Gramàtica del català contemporani: Volum 1 - Introducció, Fonètica i fonologia, Morfologia". In terms of the articles cited here, I'm not familiar with all of them, but it's possible that they were describing a different dialect of Catalan, and are not, in fact, incorrect for all dialects of Catalan. Vaaht (talk) 08:30, 19 August 2008 (UTC)

IMHO most usual pronunciation in Eastern Catalan are poble ['pɔbblə] and regla ['regglə], the archetypical examples of those two rare cases of voiced oclusive between vowels or glides. Perique des Palottes (talk) 15:18, 21 April 2010 (UTC)

Are you saying that the postalveolars of Catalan are not actually alveolo-palatal as Recasens et al say? Keep in mind that the article's transcription of caixa and baixar is specific to Valencian. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 19:23, 19 August 2008 (UTC)

I am merely saying that I have an article that the book I have on Catalan phonology states otherwise with regards to what those fricatives are. If it turns out that other articles have stated the same, then they should be included as alternate sounds.

Also, when I introduced those word examples, I was commenting on word transcriptions found in the book I have, not the Valencian pronunciations, since I know they are different. I'm sorry if I was unclear.

Also, it is unclear what is meant by "frontness" and "backness" in the following quote, since "front" and "back" are usually properties of vowels, and not of consonants: "Phonetic notes: /t/ and /d/ are denti-alveolar, having both dental and alveolar contact with the tongue. /n/, /l/, and /ɾ/ are "front alveolar;" /s/ and /r/ are "back alveolar" (or postalveolar); /ʎ/ and /ɲ/ are "front alveolo-palatal"; and /ɕ/, /ʑ/, /tɕ/ and /dʑ/ are "back alveolo-palatal." The phonetic term could be advancement and retraction, but I'm not sure. Also, if you'll glance through cited article #5 for this quote, you'll notice that the article is talking about consonant clusters, and not pronunciation in isolation, which makes it unclear why this article was referenced here in the first place. (If you can't access the article for some reason, let me know and I'll upload it for you to look at.) Vaaht (talk) 01:12, 20 August 2008 (UTC)

My understanding of "front" and "back" is exactly what you've surmised. Front=advancement and back=retraction.

It's odd, I have a pdf copy of the article but the file is corrupted and I don't recall how I was able to download it (if I recall correctly, this tidbit was actually a citation in another article that then cited Recasens and Pallares, so I could have made the citation say "cited in..."). If you could copy and paste relevent bits either here or in my talk page, I would appreciate it. Daniel Recasens seems to have done enough electropalatographic studies to determine that the postalveolar fricatives are alveolo-palatal (though he continues to transcribe them as /ʃ/ and /ʒ/, probably because it's consistent with transcription in other scholarship) as other articles I've seen by him in regards to Catalan endorse the alveolopalatal description. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 02:49, 20 August 2008 (UTC)

Just my two cents: /tʃ/, /dʒ/, /ʃ/, /ʒ/ are the habitual transcription everywhere (books, web, school etc.) for these sounds in these Catalan words: fletxa, metge, coix (although this one would be [jʃ] or [js] in Western Catalan or Valencian), puja (although this last one would be [dʒ] in most of Valencian). Those are the same sounds you can find in these English words: chop, gin, shop, vision, respectivelly. Perique des Palottes (talk) 15:18, 21 April 2010 (UTC)

They are very similar sounds and neither English nor Catalan make any sort of contrast between alveolopalatal and palatoalveolar. Since we have several sources that clearly articulate that they're the former, we can go against such tendencies in the literature since it's probably due to typographic constraints. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 16:47, 21 April 2010 (UTC)

Now you are speculating on typography. Those sounds are the same ones in English or French, and they are represented with those same symbols everywhere, here in Wikipedia and in scholarship. So you should follow your particular crusade, pardon the expression, with the charts for those other languages. For Catalan in particular you are against the whole Catalan Wikipedia and Catalan sources for starters. Perique des Palottes (talk) 17:25, 21 April 2010 (UTC)

Unreleased stops in Catalan?

The article Unreleased stop includes Catalan as a language in which final stops are not released. However, what I have heard from on-line pronunciation files of Catalan indicate that this is not the case; there is an audible release at the end of words such as Cornet and Montjuïc. Being a native speaker of Korean and being somewhat familiar with Thai, both of which do not release final stops, I hear a clear difference from the way final [t] or [k] would be pronounced in Korean or Thai and the way they are in these sound files. Maybe the sound files do not represent native pronunciations, or maybe the claim that Catalan has unreleased stops is wrong. Which is the case? --Iceager (talk) 12:09, 5 January 2009 (UTC)

I haven't seen anything about that, but I don't have any reason to believe that the sound files aren't native or near-native pronunciations. We can remove the information from Unreleased stop— Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 17:48, 5 January 2009 (UTC)

Valencian pronunciation

Valencian is a Catalan dialect, with a different language institution and it's got its own rules and pronunciation (pretty much as Catalan). The standard dialect is not the one from Valencia city, but the dialect from the South of Valencia Province and the North of Alicante Province (Southern Valencian). /dʑ/, /tɕ/, /ɕ/ and /ʑ/ do not exist in Valencian. They are /tʃ/, /dʒ/, /ʃ/ and /ʒ/ respectively. /ʒ/ exists in few words. Example; peix blau [pe(i)ʒ blau̯], caixmir [kaʒ'miɾ]. Also, /c/ and /ɟ/ do not exit in Valencian. Only in the Balearic Islands. /v/ exists in Valencian, the Valencian phonology differenciates between /b/ and /v/. Valencian also has [β̞]. In Valencian occurs a vowel hormony, i don't know if such thing is found in any other romance language. In Valencian porta is pronounced /'pɔɾtɔ/, vora /vɔɾɔ/, pilota /pilɔtɔ/, terra /'tɛrɛ/, serra /'sɛrɛ/, tela /'tɛlɛ/, merda /'mɛɾð̞ɛ/. Valencian's open vowels /ɛ/ and /ɔ/ get further open /æ/ and /ɒ/ or /ɔ̞/. As Valencian does not use vowel heights as much as Catalan, /ə/ doesn't exist. Vowel heights occurs sometimes, with "o" (pronounced /u/, /eu̯/ and /au̯/, /w/), "e" (pronounced /a/ or /i/) and "a" (pronounced /e/). Examples of Valencian vowel heights.

Ho [u], [w] or [eu̯]. Obrir [au̯'β̞ɾiɾ/. Obert [u'β̞ɛɾt]. Tossir [tu'siɾ]. Collir [ku'ʎiɾ]. Cosir [ku'ziɾ]

Escoltar [askol̪'taɾ]. Entendre [an'tendɾe]. Sencer [san'ser].Encara [aŋ'kaɾa].

Genoll [gi'noʎ]. Menjar [min'dʒaɾ]. Senyor [si'ɲoɾ].

Ell cantaria [kanta'ɾie]. Ell canta ['kante]. Ell cantava [kan'tave].

In Valencian some words with "qua-" are pronounced /ko/ pronounced

Quaranta [ko'ɾanta], quallar [ko'ʎaɾ].

Important links for the Valencian pronunciation

http://www.avl.gva.es/PDF/GNV.pdf

http://www.avl.gva.es/PDF/Diccionari/Oral.pdf

Catalan has two main dialects, Catalan and Valencian (two languages institutions for one language). Just as Portuguese, where exists a European Portuguese and a Brasilian Portuguese, however Catalan is a minority language whereas Portuguese is a global language. Would be perfect to differenciate Valencian phonetics from Catalan phonetics. 86.180.91.251 (talk) 22:19, 25 January 2010 (UTC)

Thanks for the sources. I should mention two things

1. This article does mention Valencian. I suppose there can be more added as long as it's sourced
2. Several editors (mostly in my talk page) have questioned this article's use of the alveolo-palatal fricatives when a number of reputable sources use the palato-alveolar characters to transcribe the postalveolar consonants. Because it seems that the latter set are more typographically available and because Recasens (2001), which makes the phonetic character of the postalveolars clear, it seems that even when sources use the palato-alveolar characters that we should understand them to actually be alveolo-palatal. This includes sources on Valencian.

I can't read Valencian, btw, so I'm not sure what it all specifically says. The information may be more appropriate at Catalan orthography. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 00:10, 26 January 2010 (UTC)

As it has got its own language institution (Acadèmia Valenciana de la Llengua); it specifies its pronunciation, and it's got its own phonology, grammar and orthography.

Yes, i know, but the Valencian phonology does not mention the existence of schwa /ə/, and /dʑ/, /tɕ/, /ɕ/ and /ʑ/ on its phonology inventory (as the Institut d'Estudis Catalans does) but mentions a vowel harmony. Recasens assumes Valencian pronuncation as the Catalan one, but i have never met a Valencian who pronounces /dʑ/, /tɕ/, /ɕ/ and /ʑ/ and /ə/. May be the Balearics and the Catalans would pronounce more like that, i've never really paid attention on it. 86.180.91.251 (talk) 12:36, 26 January 2010 (UTC)

I believe you about the existence of schwa (the article already implies that). Regarding alveolo-palatal fricatives, however, you'll have to excuse my scepticism; because e.g. [ʃ] and [ɕ] are acoustically and articulatorily similar and because very few languages contrast them, I doubt that many people without training can even perceive a difference. If you don't bring a source to the table, there's no reason to think otherwise. There is no assumption on Recasen's part. Indeed, Recasens & Espinosa (2007) say specifically of both Majorcan and Valencian that "the sounds traditionally labeled /ʃ/ and /tʃ, dʒ/ are alveolopalatal" (p. 143) and provide a detailed electropalatographic study of the sibilants in both dialects. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 21:35, 26 January 2010 (UTC)

Dubious

Palatal signs

All references consistently use the same palatal symbols as used for other languages' sounds, as in English, French, Spanish or Italian charts. This article should be consistent with the whole Catalan Wikipedia and Catalan sources. Assertion on palatal sounds and sign usage is not found in any of those cited sources:

• Carbonell, Joan F.; Llisterri, Joaquim (1999), "Catalan", Handbook of the International Phonetic Association: A Guide to the usage of the International Phonetic Alphabet, Cambridge: Cambridge University Press, pp. 61–65, ISBN 0-521-63751-1
• Recasens, Daniel and Pallarès, Maria Dolores (2001). "Coarticulation, Assimilation and Blending in Catalan Consonant Clusters". Journal of Phonetics 29(3): 273–301. doi:10.1006/jpho.2001.0139.

Does not seem quite believable... Perique des Palottes (talk) 18:10, 21 April 2010 (UTC)

Are you reading the literature or just looking at the symbols? Sources consistently say "alveolopalatal" even when they use the symbols for palato-alveolar. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 03:22, 22 April 2010 (UTC)

If general usage of all scholarship, and native speakers of both Valencian and Eastern Catalan, insists that those sounds are the same analogous of English, Spanish, Italian or French and insists in using those signs, then you and Recasens are in minority and if anything should expose your view as an alternative to mainstream. Perique des Palottes (talk) 07:24, 22 April 2010 (UTC)

Native speakers of French, Spanish, Catalan, Italian, or English are not equipped to tell the difference between [ɕ] and [ʃ]. They are very similar sounds. Even sources that transcribe this sound with <ʃ> describe it as alveolopalatal (i.e. [ɕ]). Recasens is the only scholar I know who has made detailed phonetic measurements of these sounds in Catalan. If you know of another phonetician who has done the same and come to a different conclusion, then by all means bring that to this discussion but we're not going to discount Recasen's work (which is cited in, e.g. Wheeler) just because it goes against your a priori understanding.

I don't have ready access to Recasen's first article that shows this ("The articulatory characteristics of palatal consonants", 1990) but according to Google scholar, this is cited in over 43 different works. That doesn't sound like a minority. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 17:50, 22 April 2010 (UTC)

What we really need is someone who speaks both Catalan and Romansh. Romansh, unlike most Romance languages, contrasts /t͡ʃ/ and /t͡ɕ/. _{(suoı̣ʇnqı̣ɹʇuoɔ · ʞlɐʇ)} nɯnuı̣ɥԀ 22:59, 24 February 2014 (UTC)

Diphtongs in variation

On Catalan rising diphtongs and afluixar: Catalan rising diphthongs (that is, starting with [w] or [j]) are only possible in these contexts:

• In word initial position, eg.io-gurt
• Just after another vowel, that is, sequence must be vowel + [j],[w] + vowel, eg. ve-uen, ve-ien
• Only [w] just after /g/ or /k/, that is, sequence must be /g/,/k/ + [w] + vowel, eg. aigua, quan

So it cannot be possible that:

"Catalan possesses two sets of diphthongs in variation; [wi] varies with [uj] (as in afluixar [aflujˈɕa~aflwiˈɕa] 'to loosen') and [iw] with [ju]."

And the example cited afluixar, must be either [afluj'ʃa(r)] or [afluj'sa(r)] (Western and Valencian), or [əflu'ʃa(r)] (Eastern Catalan or Balearic, no diphtong whatsoever). Perique des Palottes (talk) 15:18, 21 April 2010 (UTC)

That may be the case in your dialect, but this is talking about interdialectal variation, which may not be clear with the present wording. Does knowing that make it more believable? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 16:45, 21 April 2010 (UTC)

With current wording what it says is plain wrong, in any dialect of Catalan. Please, before undoing each other edits, proceed to redo the wording to make it more believable. Perique des Palottes (talk) 17:37, 21 April 2010 (UTC)

This is what Wheeler (p. 105) says: "In the case of post-tonic [iu] sequences, such as pòdium 'podium', Màrius (given name), the constraint ranking in (44) will favour diphthong formation over hiatus. Recasens remarks (1993: 107, 111) that it is hard to be certain whether the output is actually [ju] or [iw] in such cases." So this is apparently not interdialectal. I don't have access to the source that's cited (Carbonell & Llisterri 1999), but the 1992 article in JIPA that it's based on doesn't mention it. Perhaps we should try to get ahold of the 1993 edition of the Enciclopèdia Catalana. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 03:51, 22 April 2010 (UTC)

/ɛ/ and /e/

Central and Western varieties differ in their incidence of /e/ and /ɛ/, with /e/ appearing MUCH more frequently in Western Catalan. I think this is not the proper statement adjective, /e/ appears more frequently in Western Catalan (not MUCH more frequently, anyone can learn the different spelling and pronunciation rules). JaumeR (talk) 16:22, 21 July 2010 (UTC)

That's a sourced statement. Do you have a source? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 16:32, 21 July 2010 (UTC)

Nope.

Replaced "MUCH more frequently" by "MORE frequently". Jɑυмe (xarrades)

Phonemic status of affricates

Catalan is a pluricentric language, therefore this article needs to be unbaised to any dialect, mentioning all dialects' features.

• /v/ shouldn't be in parentheses.
• Affricates/Fricatives:

I don't think the phonemic status of /dʒ/ and /tʃ/ is dubious. They are "affricates" in most Western Catalan dialects, including many Eastern Catalan dialects and idiolects (esp. those where affricates geminate).

Standard Valencian phonology, much like North Western Catalan, includes a larger usage of affricates /tʃ/ and /dʒ/; occurring word-initially, intervocalically and in word-final position. The old pronunciation of j/g /dʒ/ (phonemic) is kept in Valencian (/ʒ/ everywhere else); but also, in some cases, in North-Western Catalan and southern Catalonia. Similarly, in western dialects, [tʃ] often occurs as a fortified x /ʃ/, while intervocalic ix /ʃ/ might be regarded as an allophone [jʃ] by some authors (unlike Eastern Catalan where /ʃ/ is phonemic). In the same way, /ʒ/ is an allophone in most Valencian dialects.

Concerning /ts/ and /dz/ (IMO, the only dubious phonemes) only exist between vowels and /ts/ word finally.

In many dialects all affricates are geminated in intervocalic position (doesn't this mean affricates are not clusters in these dialects?)

• What about mentioning final obstruent devoicing, as occurs in German and Dutch.
• Add some vowel charts; stressed vs unstressed vowels in Central Catalan and Valencian (Balearic could also have its own vowel chart with stressed /ə/). The vowel /ə/ generally corresponds to /ɛ/ in central Catalan, add also it corresponds to /e/ in Valencian and north-western Cat. (same as it was in Vulgar Latin, /e/).
• I think gemination should be included in prosody, along with stress and secondary stress, rather than mentioning it everywhere on the article.
• It is somehow useless the chart with diminutives, e.g. malalt - malaltet, hort - hortet.
• The consonant chart should be closer to other languages; Italian, Portuguese, French, etc. It is pointless to mention voiced and voiceless consonants.
• In Catalan occurs synaloepha (syllable merge); e.g. hi ha, posa-hi, onze anys. Nothing is mentioned about such phonological process. As you probably know, in Catalan/Valencian occurs normal elision with the apostrophe, as in French, Italian or English; e.g. l'illa, l'île, l'isola, I'm. Nonetheless, in Standard Valencian and Catalan there are many cases where the apostrophe isn't used, but occurs a sort of sandhi (i.e. hiatus either contract or diphthongise):
  • mitja hora [ˌmidʒ‿ˈɔɾa] (Italian mezz'ora)
  • una hora [ˌun‿ˈɔɾa], una escala [ˌun‿asˈkala], una amiga [ˌun‿aˈmiɣa] (Italian un'ora, una scala, un'amica)
  • quinze anys [ˈkinz‿ˈaɲs], mà esquerra [ˈma‿sˈkɛra], no es coneixen [ˈno‿s koˈnejʃen] (synaloepha-contraction).
  • no ho sé [ˈno‿w ˈse], no hi ha [ˌno ˈj‿a] (synaloepha-diphthongisation)

You can check vowel contact (vocals en contacte) on page LXXIII (14/16). [3] JaumeR (talk) 01:37, 14 October 2010 (UTC)

As the lede says, the article does try to view Catalan as a system of dialects rather than a standard language with deviating dialects.

/v/ should be in parentheses because it isn't present in many dialects.

The diminutives chart shows how consonants that aren't pronounced in words can still be there underlyingly. It also shows vowel alternations. That is certainly not useless.

As the article states, the dubiousness is partly because there are alternations between affricates and fricatives and partly because it's not clear if affricates aren't simply stop+fricative clusters. Thus, even if they appear, the academic sources vary in how they consider them.

Yes, an explicit mention of word-final devoicing is in order.

Why should we move all mentions of gemination to a single subsection of prosody? Is there something about Catalan gemination I don't know?

The reasoning for the voiced-voiceless distinction is so that we can clearly show what [c ɟ] are allophones of; putting [ʃ ʒ tʃ dʒ] in the postalveolar and palatal columns at the same time that they are the only postalveolar consonants doesn't make much sense.

Don't we already have a vowel chart for Catalan? Or do you mean a comparative table? That sounds like a good idea.

Do you have sources that talk about synaloepha and crasis? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 16:39, 14 October 2010 (UTC)

I don't think it is necessary to include allophones if we don't include all. Allophones can't really be added to phonological charts. Lenition is not included here, why include Majorcan allophones?

I will also change those examples with diminutives and add some better ones. JaumeR (talk) 21:29, 7 November 2010 (UTC)

I am not in favor of a purely phonemic chart here. Not only are [c ɟ] allophones of /k g/ for a number of speakers, but they're also the primary realization of the dorsal plosives (in which case, one could argue that [k g] are be allophones of /c ɟ/ in these dialects). The situation is similar with /v/. E.g. Majorcan features [v] and [w] in complementary distribution (making it a positional variant) but it's also a phoneme in other dialects.

More importantly, having a purely phonemic chart would force us to take a side regarding whether the affricates are separate phonemes or not, which would constitute a POV bias.

The source you've provided demonstrates that synaloepha is a feature of Valencian, though I was hoping for something that went into more detail about it. I guess there isn't much one can say about the phenomenon, though if we find such a source we can use it to expand mention of it in the article beyond a passing reference. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 23:21, 7 November 2010 (UTC)

Then if you wish you can take off the affricates, which i think All affricates aren't dubious in all dialects; at least /dʒ/ and /tʃ/ are phonemic sounds in Valencian. I doubt about [ts] (found in loanwords, in compounds and rare words as potser or lletsó, and in word final position), and [dz] (found in intervocalic position).

It's not as though making the table more than a purely phonemic one opens a floodgate that would force us to put every allophone. In this case, the affricates, palatal plosives, and v are easily put in without making the table overcomplicated. In each case, there is a clear and compelling reason to put them in an otherwise common-core phonemic table. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 01:45, 8 November 2010 (UTC)

I just added few tables. Jɑυмe (xarrades)

Recasens & Espinosa?

All Valencians don't lenite /b d g/ in utterance-final position.

I think we could delete or modify that statement as can cause confusion, all sources for Valencian say these consonants should be pronounced as plosives. Jɑυмe (xarrades) 21:16, 23 December 2010 (UTC)

"Should" be pronounced or "are" pronounced? There's only a certain amount of relevance that prescriptivists have in linguistic description and this would be an example of where they wouldn't be relevant. If it's the latter, I'm curious which sources you mean. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 09:42, 9 January 2011 (UTC)

I think D. Recasens and A. Espinosa generalise a bit too much. Personally, I've never heard such pronunciations in my surroudings (southern Valencia).

Well, according to other sources (e.g. Joan Veny), it is true Valencian dialects may voice all obstruents in certain environments, i.e. voiced plosives /b d g/ may be lenited between vowels, only when they follow a word starting with a vowel (or occlusive C) phoneme: fred àrtic, would be [ˈfɾɛt ˈartik] in Standard Catalan and [ˈfɾet ˈartik] in Std. Valencian, but [ˈfɾed ˈaɾtik]~[ˈfɾeð ˈaɾtik] in Valencian dialects, similarly bolígraf antic could be [boˈligrav anˈtik] or sac ample [ˈsag ˈample]~[ˈsaɣ ˈample] (Joan Veny, p. 161). However, in utterance-final position plosives (incl. all obstruents) normally remain voiceless [p t k].

Sources:

• Estudis de geolingüística catalana (1978).
• Els parlars catalans (1978).
• Introducció a la dialectologia catalana (1986).
• Atles lingüístic del domini català (ATLC) [Antoni M. Badia i Margarit and Germà Colón i Domènech] on which Joan Veny also contributed. Jɑυмe (xarrades) 23:07, 11 January 2011 (UTC)

I will move lenition of coda plosives to phonotactics, since this is irrelevant to both Standard Catalan and Valencian. Reasons:

• Such lenition shouldn't be in the coda, but when the following word starts with a vowel, e.g. fred àrtic could be pronounced as [ˈfɾeˈd aɾtik] ~ [ˈfɾeˈð aɾtik] in regular speech, while it will be [ˈfɾeˈt aɾtik]/[ˈfɾɛˈt aɾtik] in Standard Valencian and Catalan.
• Also, the unofficial (not accepted by the AVL and IEC) alternative Valencian spelling, known as Normes del Puig and created by the Valencian association Real Acadèmia Valenciana de la Cultura, spells all final plosives with "voiceless" letters: verb, Jacob, calb, àrid, aràcnids → verp, Jacop, calp, àrit, aràcnits. So, if Recasens statement was 100% true, why would the unofficial Normes del Puig spell words like this? :S Jɑυмe (xarrades) 23:11, 15 March 2011 (UTC)

I looked over the sourced statement and have edited the article to include a caveat that I hope you find appropriate. I wouldn't want to move this statement to phonotactics since contextual allophones are not part of phonotactics. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 03:19, 16 March 2011 (UTC)

It is perfect. Jɑυмe (xarrades) 15:09, 16 March 2011 (UTC)

Giner? Ferrer?

Jaume, this is an incomplete citation. Can you provide a full citation in the bibliography section? If you have difficulty with the citation template, just give it your best shot. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 09:42, 9 January 2011 (UTC)

Thanks, I'll add more details about these references and i'll get some sources for the assimilation. Jɑυмe (xarrades) 23:07, 11 January 2011 (UTC)

Map

Dialectal map of Catalan
Eastern dialects:
█ Northern Catalan
█ Central Catalan
█ Balearic
█ Alguerese
Western dialects:
█ North-Western Catalan
█ Northern Valencian
█ Central Valencian
█ Southern Valencian

Jɑυмe (xarrades) 19:30, 14 January 2011 (UTC)

I think this map is much better to show dialects, we could just use it as an image without citing Wheeler, seen it is not provided with a reference of this author.

It's unsourced. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 19:45, 17 January 2011 (UTC)

/y/ in Northern Catalan

According to Recasens, this vowel phoneme appears in recent loanwords in Northern Catalan (where there is a higher influence from French&Occitan). So we can mention /y/ is a marginal vowel in this dialect. See Fonètica descriptiva del català: assaig de caracterització de la pronúncia del vocalisme i consonantisme del català al segle XX. Jɑυмe (xarrades) 22:55, 2 March 2011 (UTC)

Vowel chart

Fonètica descriptiva del català: realitzacions fonètiques sistemàtiques de les vocals catalanes, Daniel Recasens i Vives

Vowels of Catalan

	Front	Central	Back
Close	i		u
Close-mid	e		o
Mid	(e̞)	(ə)	(o̞)
Open-mid	ɛ		ɔ
Open		a  ~  (ɑ)

Stressed vowels

• Most Catalan dialects contrast 7 stressed vowels /a ɛ e i ɔ o u/. Nevertheless; Balearic dialects (e.g. Majorcan), add an 8th stressed vowel (/ə/). Northern Catalan, Alguerese and some local dialects, or idiolects, influenced by the neighbouring languages have a 5- (or 6-)vowel system (i.e. mid-open and close-mid vowels may merge into mid vowels).
• Northern Catalan (rossellonès) may add two rounded vowels [y] and [ø̞] from French loanwords.
• The vowel /a/ is central [ä ~ ɐ] in standard Central Catalan, while it is front (anterior) [a ~ æ] in Majorcan and Minorcan.

Unstressed vowels

In Eastern Catalan dialects (excepting most of Majorcan) vowels reduce to 3 in unstressed position:

• [ə i u] in Central Catalan, Northern Catalan, Minorcan, and Eivissan
• [a i u] in Alguerese
• [ə i o u] in Majorcan

In Western Catalan dialects vowels reduce to 5 in unstressed syllables:

• [a e i o u] (in some cases some unstressed vowels may merge into different realisations). Jɑυмe (xarrades)

So far so good, but I don't think we need to put a whole extra row for the mid front and mid back vowels. Mentioning them in prose is enough. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 03:57, 4 March 2011 (UTC)

Something like this?

Vowels of Catalan

	Front	Central	Back
Close	i		u
Close-mid	e		o
Mid		(ə)
Open-mid	ɛ		ɔ
Open		a  ~  (ɑ)

Jɑυмe (xarrades) 07:38, 4 March 2011 (UTC)

I was thinking more like this:

Vowels of Catalan

	Front	Central	Back
Close	i		u
Close-mid	e	(ə)	o
Open-mid	ɛ		ɔ
Open		a

The difference with /a/ is also a possibility, though I could go either way on having ɑ actually in there. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 13:01, 4 March 2011 (UTC)

It looks good to me. Well, then we can use both /a/... We already contrast Majorcan /k g/~[c ɟ]. What do you think? :) Jɑυмe (xarrades) 18:25, 4 March 2011 (UTC)

You're right. It should be clear to readers that these are different representations/realizations of the same phoneme. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 21:22, 4 March 2011 (UTC)

Well, the vowel chart on Recasens is not fully right, since Majorcan/Valencian's main realization of /a/ is further front than in the Central dialects.

Vowels of Catalan

	Front	Central	Back
Close	i		u
Close-mid	e	(ə)	o
Open-mid	ɛ		ɔ
Open	(a̟)  ~  a  ~  (ɑ)

How should this be displayed? As [a]~[ä] or [a̟]~/a/? Jɑυмe (xarrades) 21:32, 9 April 2011 (UTC)

I've edited the article by simply showing a plain a in the table and letting the article's prose elaborate on the issue. Do you think it's sufficiently clear? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 23:02, 9 April 2011 (UTC)

It is, thanks.

There is a source (p.14) that mentions Majorcan & Valencian /ɛ/ approaches to [æ] and /o/ to [ɔ̞ ~ ɒ], and unstressed [ə] tends to [ɐ] in the Barcelona area. Jɑυмe (xarrades) 22:58, 11 April 2011 (UTC)

Stop deleting absurdities

• On the Spanish phonology the consonant chart is in the center. Why are you changing it?! You don't like it in the center? Well, I DO ;)
• All or most phonologies display WP:IPA for... instead of Wikipedia:IPA for..., see Spanish phonology or Portuguese phonology.
• Stressed monosyllabic words have stress (to differentiate them from unstressed monosyllabics): mon [mun] vs món [ˈmon], es [əs] vs és [ˈes], etc.
• The rest is very good, thanks for your help Aeusoes :) Jɑυмe (xarrades) 18:31, 24 March 2011 (UTC)

Yeah, the first time I took out the stress I had forgotten about unstressed monosyllables. I can't really speak for the other changes, since they were done by User:Δ. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 23:20, 24 March 2011 (UTC)

/bl/ and /gl/

Here (see conventions) says in other dialects (it only means Valencian, and perhaps La Franja dialects), (intervocalic) /bl/ and /gl/ do not geminate. I don't know right now which reference explains this (Veny? Recasens? Badia i Margarit?). I'll have a look on my books. Shall we use this until I find a better source? Jɑυмe (xarrades) 18:53, 24 March 2011 (UTC)

If you're planning on finding an adequate source anyway, what's wrong with keeping it marked as uncited until then? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 23:27, 24 March 2011 (UTC)

You are right, until then we can have it unsourced. Jɑυмe (xarrades) 00:55, 25 March 2011 (UTC)

Vowel reduction in diphthongs

The article currently doesn't mention this, but I presume that vowels are reduced the same in diphthongs as they are as single vowels? And if so, then perhaps they are not really diphthongs but phonemically sequences of vowel + approximant. I also wonder what happens when /ɔw/ or /ow/ are reduced in this way in Central Catalan (since /uw/ is not permitted). CodeCat (talk) 22:18, 25 December 2012 (UTC)

Yes, these are only phonetic diphthongs. I'm not sure how vowel reduction is treated, though. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 23:37, 27 December 2012 (UTC)

Are the rising diphthongs allophonic?

I get the feeling that rising diphthongs like /je/ and /we/ are really allophonic in Catalan. As noted above in the section "Diphtongs in variation", they only appear in a very restricted set of environments. And it seems that in these environments, the conversion from an unstressed /i/ or /u/ to /j/ or /w/ is automatic. That is, the combination /gua/ automatically becomes /gwa/, while /eia/ becomes /eja/, and word-initial /io/ becomes /jo/. It also seems that in sequences of two high vowels, the character of the diphthong is ambiguous and can be either rising or falling. If all of this is correct, then apparently /w/ and /j/ are phonemic only in the syllable coda, and allophonic otherwise. CodeCat (talk) 20:14, 22 May 2015 (UTC)

Is [ŋ] a phoneme?

In Central Catalan, final /nk/ is assimilated to /ŋ/, so that makes it contrastive at least in this position. But is it phonemic in any other cases, or in dialects that don't drop final /k/? CodeCat (talk) 14:30, 25 May 2015 (UTC)

It might be a bit OR to assume that the velar nasal is phonemic just because the surface realization deletes the /k/. Take a look at this presentation by Daniel Silverman regarding the use of alternations when it comes to phonemicity. If a Catalan word ends in [ŋ] but a suffix changes it to [ŋk], I'd argue that there still wouldn't be evidence of a phonemic velar nasal. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 16:36, 25 May 2015 (UTC)

It's true that a Catalan speaker can always "recover" the /k/ because the rule can be reversed: final [ŋ] > medial [ŋk]. But this doesn't work for the other final deletion rules like [nt] > [n]; there is no reverse rule for final [n] > medial [nt]. So the loss of [t] is phonemic and contrastive in this case. CodeCat (talk) 16:59, 25 May 2015 (UTC)

But what does that have to do with the case of /nk/~/ŋ/? --JorisvS (talk) 18:34, 25 May 2015 (UTC)

The article actually has a table outlining how suffixation shows the deletion rule to be isochronic, rather than diachronic. Thus we have banc [ˈbaŋ] ('bank') but banca [ˈbaŋkə] ('banking') and punt [ˈpun] ('point') but punta [ˈpuntə] ('tip'). — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 20:34, 25 May 2015 (UTC)

Maybe so, but that has little to do with phonemicity. The same phenomenon appear more widely in French, and in languages with word-final devoicing. The "underlying" structure in all of these languages cannot be predicted from a single phonemic word, but can only be restored by comparisons with related forms. Thus, it belongs to morphophonology and not phonology proper. Punt is still phonemically /pun/ and not /punt/, no matter how you explain it. A Catalan speaker would not be able to tell the difference between pun and punt without having prior knowledge of the morphological structure of the words. Just as a Dutch speaker can't tell whether /ɦɑrt/ is hart or hard.

However, it seems that [ŋ] is an exception. It's possible for a Catalan speaker to recover the phonemic structure /nk/ because [ŋ] only occurs in this case, and appears in complementary distribution with [ŋk] elsewhere. CodeCat (talk) 22:09, 25 May 2015 (UTC)

Phonology often gets into cases where the surface realization creates a merger or a minimal pair that is analyzed as representing a different underlying structure (which is what transcriptions between slashes are intended to represent, by the way). This is how we understand the alveolar flap to be an allophone of both /t/ and /d/, rather than another phoneme, even when American English speakers can't tell the difference between ladder and latter. This is a common analysis of Dutch final devoicing, as well. In that sense, punt can indeed be analyzed as /punt/.

This is not to say that it's the best analysis, but the point I'm making is we'd need sources to confirm an analysis of a velar nasal phoneme, otherwise we're guilty of WP:SYNTH. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 22:39, 25 May 2015 (UTC)

You're still confusing phonology with morphophonology. Phonemic structure deals only with the question "which distinctive sounds is this word made up of", and treats each word in comparison with all others in the language, but does not consider relationships between words. Morphophonology deals with the relations between different forms, and includes the concept of an "underlying form" with rules that produce a surface phonemic form from an underlying form. So the underlying structure of punt is indeed |punt| (note the vertical bars) but the phonemic structure is still /pun/; the morphophonology of Catalan includes a rule that eliminates the phonemic contrast between underlying word-final n and nt, so that both appear phonemically as /n/. Compare this to the merger of unstressed vowels; these also cause a loss of phonemic distinction. In this case, the original underlying vowels are often not even recoverable, as in many cases there are no related forms in which they appear stressed. CodeCat (talk) 23:11, 25 May 2015 (UTC)

That sort of segmentation between the two fields is not universally shared by modern linguists. I'm sure there are linguists who would account, for example, for Dutch final devoicing with the interpretation that hard is phonemically /ɦɑrt/ but morphophonemically |ɦɑrd|, but I can't say I've come across any. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 01:33, 26 May 2015 (UTC)

That's an appeal to authority. --JorisvS (talk) 07:49, 26 May 2015 (UTC)

No more than any other request for citations. Hardly illegitimate for a project grounded in verifiability. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 15:39, 26 May 2015 (UTC)

But you didn't ask for a citation. Instead, you responded to CodeCat's argument by referring to other people. Besides, discussions on talk pages can include uncited arguments (they usually do). --JorisvS (talk) 16:13, 26 May 2015 (UTC)

My point has been that whether we consider the velar nasal to be a separate phoneme depends on analysis and, if we consider this discussion to have bearing on article content, my reference to our policy on no original research would be an implicit request for citations. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 17:21, 26 May 2015 (UTC)

If no citation can be found for the result of this discussion, then of course it can't go in the article. However, that does not mean it addresses CodeCat's argument. Can we distinguish between different levels of correctness in those analyses in some way? What if a Catalan speaker is presented with [punt] in a sentence (by a native speaker), how would this be interpreted? --JorisvS (talk) 18:17, 26 May 2015 (UTC)

I can guess, but I'd have no way of really being sure. Maybe the speaker would consider it a spelling pronunciation or hyperformalism of some sort. They could also hear it as resembling a dialect that doesn't feature this sort of deletion. Or they might not notice it at all. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 18:23, 26 May 2015 (UTC)

To prevent an interpretation as it coming from some (other) dialect, the speaker would have to speak the same dialect as the listener. If they wouldn't notice the insertion at all, that would suggest to me that they are phonemically the same and hence evidence that [pun] can phonemically be considered /punt/. --JorisvS (talk) 18:33, 26 May 2015 (UTC)

Besides /nˠ/ (ŋ) and /nʲ/ (ɲ), /l̪ˠ/ and /n̪/ can also be found in the coda in many dialects (including Valencian and Balearic). Thus, a broad transcription of punt could also be /pun̪/. IMO this type of transcription reflects better a pan-dialectal pronunciation rather than /punt/ or /pun/. — Jɑuмe ^(dis-me) 14:16, 18 January 2016 (UTC)

A diaphonemic approach to transcriptions in Catalan sounds interesting, though I don't know if it makes sense for us to pursue it in our own transcriptions. Do you know if there are sources that approach the matter? I vaguely recall that Wheeler does something like this for Occitan in The Romance Languages. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 16:09, 18 January 2016 (UTC)

I'm not too sure if there is such approach yet, however we have a fair amount of sources that describe our dialects. Regarding [n̪], it is found on Recasens book — Jɑuмe ^(dis-me) 16:40, 26 January 2016 (UTC)

Proposals

1)

I propose:

• the creation of a page for Valencian phonology, since this article is centered on Central/Eastern Catalan, e.g.
  • [buˈliɣɾav buˈnisim] instead of [boˈliɣɾav boˈnisim]
  • Wheeler (1979) and Mascaró (1976) analyze Central Eastern varieties, the former focusing on the educated speech of Barcelona and the latter focusing more on the vernacular of Barcelona, and Recasens (1986) does a careful phonetic study of Central Eastern Catalan.[1][2]
• the creation of a chart for dialectal transcriptions based on Recasens and Veny's books (this could be very useful for users interested in dialectal pronunciations)
• the creation of an {{IPAc-ca}} template to add transcriptions from spelling. — Jɑuмe ^(dis-me) 16:40, 26 January 2016 (UTC)

I don't think a separate page is necessary. We have, for example, Serbo-Croatian phonology and Hindustani phonology. If you feel like the focus is untowardly on Central/Eastern, we can address that by adding more on Valencian.

By "chart for dialectal transcriptions" do you mean like correspondences between cognate words in different dialects? That sounds interesting. It might go better at Catalan dialects.

If you can create an {{IPAc-ca}}, I say go for it. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 17:51, 26 January 2016 (UTC)

I don't think it's professional to (randomly) mix varieties in a single article, as it doesn't help users to choose one variety when they're learning/reading our language. Besides that, Catalan is pretty much like English, with RP (Standard British English) and GA (Standard American English) (and you have a phonology article for those varieties and others like North American English regional phonology). Other reasons to create the Valencian phonology are:

• The vowel chart excludes Valencian
• The phonology section (in the Valencian article) might become too large and might confuse readers that are interested in other aspects of Valencian
• I don't like double transcriptions like [buˈliɣɾav buˈnisim] ~ [boˈliɣɾav boˈnisim]

By chart for dialectal transcriptions I mean something like this:

• International Phonetic Alphabet chart for Catalan dialects

I will try to do it when I finish other projects — Jɑuмe ^(dis-me) 21:40, 26 January 2016 (UTC)

It's not really our place to help readers learning to speak Catalan. This is a report about the sounds of Catalan, not a language-learning guide. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 01:42, 27 January 2016 (UTC)

I don't pretend that. The creation of this guide should be based on several sources that point the existence of different vowel systems with different vowel qualities in our dialects and different consonantal alternations and mergers (like the set of sibilants or /b/ and /v/) — Jɑuмe ^(dis-me) 04:39, 17 February 2016 (UTC)

2)

I also propose:

• to use a broad transcription for Standard pronunciations
• to use narrow transcriptions for our dialects
• to distinguish between Standard and Vernacular (Central) Catalan
  • to add a vowel chart for all Catalan dialects and one for central Catalan only
  • keep the vowel chart of the triad with the examples sac, sec/set, séc, soc, sóc, sic and suc

— Jɑuмe ^(dis-me) 23:31, 26 January 2016 (UTC)

But we need sourced vowel charts. Are you aware of sources that provide the vowel charts you want? Peter238 (talk) 18:26, 28 January 2016 (UTC)

We have several sources — Jɑuмe ^(dis-me) 19:58, 28 January 2016 (UTC)

You can e-mail me the scanned vowel charts and I'll vectorize them and upload them here. Peter238 (talk) 16:52, 17 February 2016 (UTC)

Done. I sent you the Valencian vowel chart first, the other dialects have imprecise charts with dense wording. Majorcan, for example, combines Valencian and a phonological /ə/ — Jɑuмe ^(dis-me) 03:13, 18 February 2016 (UTC)

Final -r

The article currently says next to nothing about the elision of final -r. Where does it occur, which dialects preserve it? Is it a predictable phenomenon, or are there exceptions? What about final -r in a cluster? CodeCat (talk) 22:01, 3 August 2016 (UTC)

In general, final /r/ is elided in Central Catalan, Roussillonese, Balearic, Algherese and Northwestern Catalan. It is preserved in Valencian, except in two geographic areas located in the north and south of this dialect. In the north final /r/ is not preserved in the comarques (districts) of e[l]s Ports, el Baix Maestrat, l'Alt Maestrat, l'Alcalatén and partially in la Plana Alta. In southern Valencia it is omitted or fluctuates in les Valls del Vinalopó, el Baix Vinalopó, l'Alacantí, l'Alcoià and la Marina. Moreover, in many Valencian dialects (including those where /r/ is usually preserved), some morphological combinations can be refractory/reluctant to maintain final /r/.

It has been suggested that final /r/ elision in Old Catalan had originated from an assimilatory process -rs → -s (Coromines). Thence the reason why /r/ is universally elided in diners /diˈnes/.

As far as I'm concerned it is predictable, although there are numerous local exceptions.

Final /r/ in clusters (mostly found in Balearic and Algherese) is preserved as either [ɾ̩] or [ɹ̩]. I hope my explanation helps. — Jɑuмe ^(dis-me) 07:41, 30 November 2016 (UTC)

Phonemic charts

I just added few loan phonemes to the consonant chart.

I would also suggest to use Recasens' pan-dialectal vowel chart and add some extra dialectal and loan vowels. — Jɑuмe ^(dis-me) 15:31, 3 March 2017 (UTC)

Vowels of Catalan

	Front		Central		Back
Close	i   (y)				u
Close-mid	e	(ɛ̝)   (œ̝)	(ə)		(ɔ̝)	o
Open-mid	ɛ					ɔ
Near-open / Open	(æ)				(ɒ)
	a  ~  (ɑ)

External links modified

Hello fellow Wikipedians,

I have just modified 5 external links on Catalan phonology. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20110814193449/http://dlc.iec.cat/results.asp?txtEntrada=tsar to http://dlc.iec.cat/results.asp?txtEntrada=tsar
• Added archive https://web.archive.org/web/20110814193526/http://dlc.iec.cat/results.asp?txtEntrada=tsuga&operEntrada=0 to http://dlc.iec.cat/results.asp?txtEntrada=tsuga&operEntrada=0
• Added archive https://web.archive.org/web/20110814193534/http://dlc.iec.cat/results.asp?txtEntrada=txec to http://dlc.iec.cat/results.asp?txtEntrada=txec
• Added archive https://web.archive.org/web/20100930153153/http://www2.iec.cat/institucio/seccions/Filologica/gramatica/fonetica/05diftongstriftongshiats.pdf to http://www2.iec.cat/institucio/seccions/Filologica/gramatica/fonetica/05diftongstriftongshiats.pdf
• Added archive https://web.archive.org/web/20110529085634/http://www2.iec.cat/institucio/seccions/Filologica/gramatica/default.asp to http://www2.iec.cat/institucio/seccions/Filologica/gramatica/default.asp

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 18 January 2022).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 02:39, 1 August 2017 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on Catalan phonology. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20100930022057/http://www.avl.gva.es/PDF/Diccionari/Oral.pdf to http://www.avl.gva.es/PDF/Diccionari/Oral.pdf

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 18 January 2022).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 09:14, 15 December 2017 (UTC)

Use of ɫ character

If [ɫ] (velar(ized) l) is just an allophone of the /l/ phoneme, why is it used so often in this article: e.g. color [kuˈɫo] ('color'); malalt [məˈɫaɫ]; bolígraf boníssim [buˈɫiɣɾəv buˈnisim], etc.??

Velar ɫ is supposed to be a syllable-final variant, yet it is syllable-final only in once in the above examples. This ɫ character appears 32 times in total!

Agihard (talk) 22:13, 17 February 2018 (UTC) Thomas Eccardt

I think that's explained in the article here. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 00:47, 18 February 2018 (UTC)

But the use of velarized l [ɫ] in these positions happens only in a minority of dialects (according to this page), and these usages are not in reference to these dialects. Currently, the page is inconsistent: sometimes it transcribes the majority dialects, other times the minority, creating confusion. I would be willing to make the page consistent with the majority dialects. Agihard (talk) 15:38, 19 February 2018 (UTC)Agihard

Yeah, that makes sense. Maybe we should default to the plain l in transcriptions unless there's a clear reason to indicate the velarization. It seems like we're doing the opposite of that now. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 17:18, 19 February 2018 (UTC)

Recasens & Pallarès 1995?

What is this Recasens & Pallarès (1995)? It can't be Recasens, Fontdevila & Pallarès (1995) because the page numbers and the topic don't match. Google is of no help either. Qwerty12302 (talk | contributions) 18:57, 10 September 2018 (UTC)

Could it be Recasens & Pallarès (2001)? — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 19:55, 10 September 2018 (UTC)

Split

I think the current article is a bit confusing for the following reasons:

1. There are two or more transcriptions for every term which makes it difficult to read and have precision
2. Some sections repeat certain info or info is mixed, for instance there is not a clear aspect of what is dialectal and what it's not (e.g. certain lenition aspects).

I suppose we could clean up the article and split Valencian (Western Catalan) from here. The map here shows a division line between East and West so we could take that into consideration to procede to do the separation of both variants. — Jɑuмe ^(dis-me) 22:54, 13 November 2021 (UTC)