Help talk:IPA/Portuguese and Galician/Archive 2

Archive 1

Archive 2

Archive 3

Nasal semivowels

While I like the idea of nasal semivowels, ȷ̃ doesn't show up right on my display and I suspect this may be common. I suspect that [ȷ] is a dotless j. Is there a problem with just using [j̃]? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 02:42, 19 January 2010 (UTC)

Yes, [ȷ] is the dotless j; I guess you don't have a font that includes it, because it looks right to me. The only problem with using [j̃] is that the dot and the tilde overlap with each other. +Angr 09:04, 19 January 2010 (UTC)

I got this symbol by copying and pasting from the article Portuguese_phonology#Consonants, where it appears twice, one saying it may be a phonetic variation of [ɲ], and the other in phonetic transcription of abdomen [abˈdomẽȷ̃], and I notice this same symbol is also used in the article pt:Língua_portuguesa#Exemplos_de_frases of the Portuguese Wikipedia, in the transcription of corações in Rio de Janeiro pronunciation, although who made the transcription of the Lisbon version used the symbol [ĩ ̯], and this latter is also used for both in the English version of this article Portuguese_language#Examples_of_different_pronunciation. --Luizdl (talk) 23:54, 19 January 2010 (UTC)

I've tried several times to remove [ȷ] from Portuguese phonology. If we're going to analyze Portuguese oral diphthongs as [j] or [w] in the syllable coda, couldn't we also do the same with the nasal diphthongs? Just assume it's a regular semivowel following a nasal vowel. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 07:53, 20 January 2010 (UTC)

As far as you are very more expert in this area than me I will respect you if you remove the [w̃] and [ȷ̃], but I can ensure you that if someone transcripts, for example, corações as [coɾaˈsõjs], without indicate the nasalization of [j], any native Portuguese speaker would think that's strange, because in the mean time they would see a phonetic representation, they wouldn't see a nasalization made by all native speakers. In my opinion we should insert an image with a /j/ with a tilde, and a footnote explaining that some fonts may not render corectly. --Luizdl (talk) 00:55, 21 January 2010 (UTC)

We could indicate here that a semivowel following a nasal vowel is nasalized. I'm not real familiar with how the literature typically represents nasal diphthongs. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 01:58, 21 January 2010 (UTC)

Good idea --Luizdl (talk) 02:27, 21 January 2010 (UTC)

Okay, I've added a note on the semivowels. Look good? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 03:12, 21 January 2010 (UTC)

Yes! By the way, what about the English examples for /j/ and /w/? some examples used are "boy" and "cow", aren't they pronounced [bɔɪ] and [kaʊ] in Enlish?--Luizdl (talk) 01:46, 22 January 2010 (UTC)

The English examples are approximations. It takes a trained ear or use of special software to notice the difference. I've been editing approximant consonant off and on lately that might explain the difference, but basically the relationship that [i] has with [j], English [ɪ] has with the final element of boy. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 01:55, 22 January 2010 (UTC)

A CHART FOR DIPHTHONGS

Yesterday I posted “THE WORD PAÍS - some comments” in the headline “Paíz” but got no feed-back. Since you answered to “A Chart for Diphthongs” so quickly, I presume I put the other post in a place not easy to notice. So I’m posting it again.

As you’ll see, the spelling “paíz” needs to be changed. Until ca. the 1930’s the spelling was paiz (no accent), and then it was changed to país. Also the table “Explanation — Stress” needs to be changed (2nd transcription of “pais”). I didn’t do it for I don’t know how far I can go when it comes to delete information…

Really, if everybody started to delete whatever they don’t agree with, I don’t know if Wikipedia could work… But then I’m probably a reliable source of information: I’m a native EP speaker, from Lisbon, and I have a degree in Linguistics. I studied 4 years at FLUL (Lisbon University) and had a scholarship for a research project in phonetics.

THE WORD PAÍS - some comments

1. Spelling. The correct spelling is país (pl.: países).

2. Pronunciation (in Portugal and some regions of Brazil). [pɐˈiz] - when the next word begins with a vowel. [pɐˈiʒ] - when the next word begins with a voiced consonant. [pɐˈiʃ] - when the next word begins with a voiceless consonant.

Actually this rule applies to all instances of s at the end of a word (or syllable) boundary:

[ʃ] – isolated word / following word begins with voiceless consonant

[ʒ] - following word begins with voiced consonant

[z] - following word begins with vowel

In some areas of Brazil s is always pronounced [s], regardless of its position in the word.

Pais ~ País. There is some inaccuracy in the transcription in the table above. In EP the pronunciation [ˈpɐjʃ] (pais) does not exist (I doubt that it exists in BP, but here I’m not 100% sure); it is [ˈpajʃ] (monosyllabic). As for país (disyllabic), in EP it is pronounced [pɐˈiʃ] and in BP [paˈiʃ]. Note the [ʃ] (isolated word).

In EP the diphthong [ɐj] is an allomorph of /ej/ in the pronunciation of Lisbon (by the way, that’s where I was born and am presently living) and some other southern regions (it tends to spread due to the prestige of the variety). /ej/ has two other allomorphs: [ej] (North) and [e] (South; note the monothongisation) Analogica (talk) 22:11, 16 February 2010 (UTC)

País (not paíz) - some comments

Yesterday I posted “THE WORD PAÍS - some comments” in the headline “Paíz” but got no feed-back. Since you answered to “A Chart for Diphthongs” so quickly, I presume I put the other post in a place not easy to notice. So I’m posting it again.

The spelling paíz needs to be changed. Until ca. the 1930’s the spelling was paiz (no accent), and then it was changed to país. As you’ll see below, the table “Explanation — Stress” also needs to be changed (2nd transcription of pais). I didn’t do it for I don’t know how far one can go when it comes to delete information…

Really, if everybody started to delete whatever they don’t agree with, I don’t know if Wikipedia could work… But then I’m probably a reliable source of information: I’m a native EP speaker, from Lisbon, and I have a degree in Linguistics. I studied 4 years at FLUL (Lisbon University) and had a scholarship for a research project in phonetics.

THE WORD PAÍS - some comments

1. Spelling. The correct spelling is país (pl.: países).

2. Pronunciation (in Portugal and some regions of Brazil). [pɐˈiz] - when the next word begins with a vowel. [pɐˈiʒ] - when the next word begins with a voiced consonant. [pɐˈiʃ] - when the next word begins with a voiceless consonant.

Actually this rule applies to all instances of s at the end of a word (or syllable) boundary:

[ʃ] – isolated word / following word begins with voiceless consonant

[ʒ] - following word begins with voiced consonant

[z] - following word begins with vowel

In some areas of Brazil s is always pronounced [s], regardless of its position in the word.

Pais ~ País. There is some inaccuracy in the transcription in the table above. In EP the pronunciation [ˈpɐjʃ] (pais) does not exist (I doubt that it exists in BP, but here I’m not 100% sure); it is [ˈpajʃ] (monosyllabic). As for país (disyllabic), in EP it is pronounced [pɐˈiʃ] and in BP [paˈiʃ]. Note the [ʃ] (isolated word).

In EP the diphthong [ɐj] is an allomorph of /ej/ (spelling ei) in the pronunciation of Lisbon (by the way, that’s where I was born and am presently living) and some other southern regions (it tends to spread due to the prestige of the variety). /ej/ has two other allomorphs: [ej] (North) and [e] (South; note the monothongisation). E.g.: leite, beijo. Analogica (talk) 22:20, 16 February 2010 (UTC)

You know that the word país/paiz doesn't appear in the guide anymore, right?

Your description of dialectal and contextual variation seems accurate to me. The issue (for me, anyway) is how this guide should reflect the dialectal variation of Portuguese. Should we pick one over the other (not a good choice)? Should we have two standards? Should we try a pan-dialectal transcription like at WP:IPA for English? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 22:30, 16 February 2010 (UTC)

First of all, let me correct two inaccuracies (And please excuse all my mistakes in English. Please feel free to correct me: as I often say, I’m an eager learner!)

1. In the transcription of país in BP, I wrote [pa’iʃ], but “s” should not be transcripted as [ʃ] – the issue was the [a’i] and I forgot the “s”.

2. “The diphthong [ɐj] is an allomorph of /ej/”. The way I put it seems that it’s only an allomorph of /ej/. That’s not correct. Though it’s not very common in EP, there are words as, say, “caiar” where you have the diphthong [ɐj].

As for the IPA chart and the dialectal variation in PT, I think you just can’t ignore the Portuguese and the Brazilian standards. In my opinion, your suggestion of a pan-dialectal transcription like at WP:IPA for English is definitely the best way to do it. I think it won’t be difficult – team work always pays!

Now, for a more accurate description of the sounds, I think we should include diphthongs, both oral and nasal (and for the latter we should use the tilde in both vowel and glide). Actually many pages of IPA for different languages include diphthongs.

And probably there’s some editing to be done. E.g. pequeno, where in BP the first “e” reads “i”. But you know, some (usually uneducated) EP speakers also pronounce it like that. But is it possible to infer some kind of phonological rule? Or is it an exception? What I mean is that we have to be careful not to get lost –not in translation- but in exceptions ;-).

I also have a doubt about “boneco”. The first “o” reads [u] in BP as well? There’s also fungo / função (ũ) and conto / ponteiro (õ). The examples for each of the two pairs are too similar: f + un; on + t. We could introduce some variety, both in phoneme (specifically stress is a key element I EP phonology) and context. I’m thinking of, say, bomba, Londres, bombom, apontar, um, atum, Hungria…

There is also an important decision to make: shall we just describe the sounds or shall we also -as it seems to be the case as it is now- provide details (and exceptions) about spelling?

If we get much into details, don’t we risk ending up with too many examples, instead of having a chart easy to read and understand? Is the chart intended to provide the tools to make a narrow or a broad phonetic transcription?

I have another suggestion. I’m a teacher of Portuguese as a Foreign Language. As a teacher of Portuguese as a Foreign Language, I prepared a general chart about EP sounds + stress + spelling to give away to my students. Though it is obviously a different approach, it could probably be of some help. Just wonder how to mail it (it’s in MS Word) Analogica (talk) 09:55, 18 February 2010 (UTC)

Okay, so you're saying that there are some words that are [ɐj] in both dialects and some words that are [ɐj] in EP and [ej] in BP. Your example was prevocalic. Are the former simply instances where the j element is resyllabified to a following vowel?

You contradicted yourself, so I'm not sure what you're advocating. Should we have method of transcription or two?

I looked through the other IPA guides and with the exception of Persian (which I've just changed) and Hebrew, all representations of falling diphthongs don't end with a semivowel but with a vowel (such as Spanish [ai]. Because of this and because the nasal j doesn't look right on a number of computers, I propose that our representation of falling diphthongs be without semivowels (though the rising ones should still be with the semivowel). This way, céu would be transcribed as [sɛu]. To distinguish between diphthongs and hiatus, we would then include the syllable boundary as something to potentially use.

Regarding exceptions, I think our examples should represent what is typical. If we feel there are notable examples, the footnotes would be the best place for this.

Is it possible to copy and paste the text of your word file into the talk page here? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 19:21, 18 February 2010 (UTC)

Analogica, about your doubt on the Brazilian pronunciation of boneco, I am Brazilian and we really have in some words the pre-stressed /o/ pronounced as [u], some other examples are coar, coador, abençoar, comida, cobiça etc.--Luizdl (talk) 23:00, 18 February 2010 (UTC)

AWB review

bringing transclusions into line with key using AWB. I'm not correcting mistranscriptions, just flagging characters that don't appear in the key and replacing the obvious ones, like <h, x, r> with <ʁ>, <au> with <aw>, etc. Please let me know if I'm forgetting anything or messing anything up. — kwami (talk) 18:21, 4 May 2010 (UTC)

I agree, but you should take some care when replacing /au/ to /aw/ for example, recently someone tried to fix the transcription in Rio de Janeiro replacing /u/ to /ʊ/, the word rio in meaning of "river" is pronounced by some as diphthong and by others as hiatus, and in the meaning of "i smile" is pronounced by everybody as hiatus making pair with (ele) riu "he smiled", pronounced as diphthongs by everybody.--Luizdl (talk) 18:48, 4 May 2010 (UTC)

Should I list them here, so that you can verify? If I'm making too many errors, I can delete that part of the coding.

As a level of redundancy, to make it clear that there is hiatus, we could use <.> Transcriptions without either . or j/w would thus be ambiguous, and await confirmation. — kwami (talk) 19:02, 4 May 2010 (UTC)

Guinea-Bissau, Diogo Cão (I assume all < ão > are diphthongs, so won't list any more; same with < em > as in Belem), Aachen (Aquisgrão), Madeira, Bartolomeu Dias, Itaipu Dam, Luís de Camões, Luiz Inácio Lula da Silva (Inacio), Battle of Vimeiro, Leiria, Queiroz,

These seem accurate for me, with exception of Aachen (Aquisgrão), that r is taped. And about what i've said above about rio, if you need source for what I said, I just found this one on Google [1], and sorry for that strange summary text on my last post, that was because in my operating system when I click with middle button it copies the selected text and paste.--Luizdl (talk) 19:23, 4 May 2010 (UTC)

Okay, the Aachen was a problem w the original. I should've caught it, but was not looking for errors. I'll be adding more words to the list above as I go along. — kwami (talk) 19:26, 4 May 2010 (UTC)

Why on earth are we giving the Portuguese name of Aachen anyway? +Angr 19:34, 4 May 2010 (UTC)

Generally the syllabic /u/ only happens when the written form is with the letter o, or when the word is derived as in "reunir" (reunite) or after a historic elided consoant as in saudar and sair (from latin salutare and salire).--Luizdl (talk) 19:47, 4 May 2010 (UTC)

Is it always the case with < o >? E.g., Nazario, Antonio, Inacio, Mario, Teodosio? Are there cases (Sampaio, Leiria, Rui Costa, etc.) where V + < i > does not become [j]? Or < e >, as in Marcelo Caetano? We don't have jV here, but have [ʒɐnju] in Jânio. — kwami (talk) 20:29, 4 May 2010 (UTC)

Words like Nazário are commonly pronounced fonetically in the three ways, ju, iw, and iu, originally it is syllabic, for that I prefer to transcript as syllabic, and because our orthography also does it in this way, the accent in orthography is avoided in paroxytones as it is in Spanish, but as words like Nazário is considered proparoxytones, it has an accent in the third last syllable, differently in Spanish, where Nazario doesn't have accent because in Spanish it is paroxytone. Sampaio and Rui are always diphthong. Caetano is also pronounced in both ways, in our dictionaries, the syllables separation in these cases are made with a colon : instead a point ., meaning that both pronunciation are possible fonetically, for that is better we transcript them as hiatuses, and let the symbols j and w only for words that is pronounced only as diphthong, like it is made in our orthography.--Luizdl (talk) 21:20, 4 May 2010 (UTC)

Okay, I'll generally leave <Vo> and <Ve> alone, and will convert but won't bother to post <Vi> and <Vu>. I think (agree?) we could mark . when it is always hiatus (except of course in rapid speech), diphthong only when it is always a diphthong, and leave it unmarked when it's variable. — kwami (talk) 21:33, 4 May 2010 (UTC)

I Agree--Luizdl (talk) 21:38, 4 May 2010 (UTC)

This one is just weird. — kwami (talk) 22:10, 4 May 2010 (UTC)

I would pronounce [eleɐ̃ˈnoɾ], but as I am Brazilian I can't say how exactly would an European pronounce.--Luizdl (talk) 22:32, 4 May 2010 (UTC)

I found the editor who transcripted it, [2], User:Joaopais, his last contributions was EP IPA transcriptions Special:Contributions/Joaopais--Luizdl (talk) 22:57, 4 May 2010 (UTC)

I've figured it out. It's "Leonor" is Portuguese, but was missing from that art. Several other "Eleanor of X" articles had it. — kwami (talk) 23:33, 4 May 2010 (UTC)

Finished. Allophones of R and L should hopefully now be inline with the key. — kwami (talk) 07:03, 5 May 2010 (UTC)

Nice work.--Luizdl (talk) 02:54, 7 May 2010 (UTC)

English approximation for ɲ

I think we should change the equivalent English approximation for instead using the [nʲ] of canyon, use the [ŋ] as in singer that sounds better to a Portuguese [ɲ].--Luizdl (talk) 02:05, 12 June 2010 (UTC)

How is that closer? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 05:09, 12 June 2010 (UTC)

The sound produced in singer resembles better to my ears, perhaps its similarity is because both are dorsal consonants, while canyon is laminal, I don't know exactly why.--Luizdl (talk) 02:16, 13 June 2010 (UTC)

It seems to me that the acoustic similarity is greater with canyon than with singer. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 02:20, 13 June 2010 (UTC)

/tʃ/ as in ‹tch›

Do we really want to include this? Do (European?) Portuguese speakers regard this as a single sound, or a sequence of /t/ + /ʃ/ in loanwords? It should be noted that French very often does the same thing (as well as ‹dj› for /dʒ/), but we don't include /tʃ, dʒ/ in WP:IPA for French, even in a "marginal" section. — ˈzɪzɨvə (talk) 07:42, 20 July 2010 (UTC)

That's a good question. I would imagine that Brazilian speakers' attitudes would be the most telling since they already have the affricate in their speech. Are they considered the same sound? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 14:21, 20 July 2010 (UTC)

The sounds [t] and [tʃ] are considered the same sound in Brazil only when before an /i/ or /ĩ/, about the word "tchau", it was borrowed initially into Brazilian Portuguese by Italian immigrants in 19th, today it does not have exactly the same meaning of italian ciao, which means in Italian both "hi" and "bye", in Portuguese it means only "bye" or the name of a farewell gesture with the hand, and is the most common and wide way of to say "bye" in Brazil. This word is also used in Portugal, I've read they learnt this word by Brazilian soap operas, but recently I read a Portuguese man to write "(t)chau" in a forum, using parenthesis, when discussing exactly about this same word.

There are some others words with tch in Brazilian Portuguese, but generally in Portugal are spelled with ch, some examples are "Tchecoslováquia" (Czechoslovakia), "República Tcheca" (Czech Republic) "tcheco" (Czech) and in gaúcho dialect, in Rio Grande do Sul, the word tchê. Another good observation is that "tchau" makes pair with "tal" (such), and between "tcheco" and "checo" (I check).--Luizdl (talk) 23:31, 20 July 2010 (UTC)

But do Brazilian speakers perceive the initial sounds of tipo and tchau to be the same? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 04:15, 21 July 2010 (UTC)

Generally no, only if you ask them the differences between the two, or in northeastern dialects, where they do not palatalize the /t/ before front vowel.--Luizdl (talk) 02:00, 22 July 2010 (UTC)

Okay, that tells me that tchau has a cluster, not an affricate. So we should take it out. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 13:54, 22 July 2010 (UTC)

But, if we already have [tʃ] in the table for "tipo", cant we just also give "tch" examples together the /t/ allophony, as they have the same pronunciation?--Luizdl (talk) 23:35, 22 July 2010 (UTC)

With "initial sounds" you meant the first phoneme or the stop? the "t" in "tipo" is what I said it generally not perceived as affricate, it is generally perceived as if it was [t], except for those speaker of northeastern region, as I've said above, the /t/ just inherit the /i/ qualities.

The tch is the affricate, else it would be pronounced [tʃiˈʃaw], as well as Djibuti is pronounced [dʒiʒiˈbuti], listen this file uploaded at forvo dated in 2009/09/09.

http://pt.forvo.com/word/djibuti#pt

--Luizdl (talk) 00:11, 23 July 2010 (UTC)

You said that tipo and tchau were perceived as different sounds by speakers who exhibit the affricate pronunciation in tipo, right? Or did you misunderstand my question? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 01:00, 23 July 2010 (UTC)

Yes, tipo and tchau are perceived as different sounds, although they have the same pronunciation, the /t/ before /i/ is still perceived as /t/, but is pronounced in quite same way of tch, although the unique reason the /t/ is pronounced as [tʃ] is because it gets palatalized by /i, ĩ/. I don't know if I understand your question, in my first answer, I understood if you was asking me if they perceive the entire sound tipo is the same of the entire sound tchau, but when you said it means tchau is a cluster, I doubted if you was asking if its initial pronunciation (the stop) of tipo is perceived as the same of tchau, because clusters tends to have epenthesis in Portuguese, with some few exceptions of complex consonants as /kw/ and /ks/.--Luizdl (talk) 01:37, 23 July 2010 (UTC)

It seems to me, if they're perceived as different sounds, that Brazilian speakers make a distinction between an affricate [tʃ] and a stop+fricative cluster [tʃ]. A cluster that could simply be another exception to the epenthesis tendency.

A parallel in English is the affrication of /t/ and /d/ before /r/. Words like drugs and trick, for a number of speakers, have the initial stop pronounced more in the postalveolar area and with some affrication. Although the precise phonetics of these allophones vary from speaker to speaker, there is some evidence that speakers perceive these affricates as the same as in the words jug and chick even though they're allophones of /t/ and /d/. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 01:50, 23 July 2010 (UTC)

You may be right, in my opinion we should keep it, but I wont revert you if you take it out from the table.--Luizdl (talk) 01:59, 23 July 2010 (UTC)

How are they pronounced?

In Northern Brazil, how are /d/ and /t/ before /i/ pronounced? Are they affricated like their counterparts in most Brazilian varieties or not? —Preceding unsigned comment added by 118.173.94.20 (talk) 20:18, 28 August 2010 (UTC)

When I listen Amazonians people at television, I notice they also affricate the /d/ and /t/, where these plosives generally are not affricated is mostly in Northeastern Brazil, many northeastern people migrate to here in São Paulo state for work, and most of them start to affricate these plosives after some time living here, but the northeastern accent is easily identified by lengthening the vowels and by opening some pre-stressed vowel.--Luizdl (talk) 03:59, 30 August 2010 (UTC)

Thank you. --125.24.90.110 (talk) 16:40, 30 August 2010 (UTC)