In most Brazilian regions the "s" sound at the end of a syllable is non-palatized and sounds exactly like "s". It may be more stressed in one place and less in another but its pronunciation tends to be more or less homogeneous. Rio de Janeiro is an exception, though. Here, the final "s" sounds like "sh", for instance, "dois" is pronounced "doish". This phenomenon was acquired when the Portuguese Court was transferred to Brazil during the Napoleonic invasion of Portugal. Fifteen thousand Portuguese aristocrats arrived in Rio de Janeiro and their habits, including their way of speaking, influenced the local people, so we keep that "sh-sounding s" at the end of syllables (and words) until today. My question is: is the palatized "sh-sounding s", as I have explained, pronounced in the whole country of Portugal or are there places where a non-palatized "s" is pronounced?
In fact, both in Portugal and Brazil the pronunciation of s at the end of a syllable depends, even if you don’t realise it, on what comes after it: it is commonly pronounced /ʃ/ (as shape or ch in the Portuguese word chato) in Portugal and Rio and /s/ (as in samba) elsewhere in Brazil if followed by an unvoiced consonant such as c, f, p, t or by a long pause; and it is pronounced /Ʒ/ (as s in decision or j in the Portuguese word jarro) in Portugal and Rio and as /z/ (as in zoo) elsewhere in Brazil if followed by a voiced consonant such as b, d, g, m, or v. Of course you don’t have to worry about this if you’re only interested in pronouncing correctly: you vocal system will correctly choose between /ʃ/ and /Ʒ/ and between /s/ and /z/ without you even thinking about it. So (only boldface /ʃ/, /Ʒ/, /s/ and /z/ are intendend to show actual pronunciation; the rest of the word is just as it is normally written):
Esbarrar is pronounced eƷbarrar in Portugal and Rio or ezbarrar elsewhere in Brazil.
Mesmo is meƷmo or mezmo.
Esvair is eƷvair or ezvair.
Casco is caʃco in Portugal & Rio and casco elsewhere in Brazil.
Casta is caʃta or casta.
At the end of a word the sound depends on the first sound of the next word unless there is a long pause, in which case it is /ʃ/ or /s/. If the next word starts with a vowel the sound is /z/ in both Portugal and Brazil.
Bons dias sounds bonƷ diaʃ in Portugal & Rio or bonz dias elesewhere in Brazil.
As portas abertas sounds aʃ portaz abertaʃ or as portaz abertas.
The reason for this is that the mouth takes the same position to pronounce /ʃ/ or /Ʒ/, the only difference being that the throat vibrates to produce /Ʒ/ (that is why this sound is called voiced) and does not vibrate to produce /ʃ/ (that is with it is unvoiced). The same is true of the voiced /z/ and unvoiced /s/. You can check this by placing your finger tips on your throat (see this answer). And the throat cannot stop or start vibrating fast enough to smoothly pronounce one voiced consonant followed by an unvoiced one or vice-versa. So we pronounce the s as voiced (/Ʒ/ or /z/) if the following consonant is voiced; and as unvoiced (/ʃ/ or /s/) if the following consonant is unvoiced
Centaurus, I understand your question. In Portugal we say "doish", it sounds like "ch". All Portuguese pronounce dois with "sh" in the end.
I used this converter and it seems that the phoneme is:
Almost; in the Trás-os-Montes dialect they pronounce it /ʒ/.
The s at end of a sylable is pronounced /ʃ/ in most of Portugal, and that is the correct pronunciation.
But in the dialeto transmontano (interior north of Portugal) the /ʃ/ is pronouced /ʒ/, like the j in gajo.
With this accent, Viseu would sound like "Vijeu".
The Wikipedia page links to a sound sample, which I recommend.
These thick accents are becoming less common, in my opinion due to the uniformization of language and the influence of TV's centralized accent.
For more information, read the WikiPedia page on dialects of Portuguese.
The subpages are not very organized, but these articles are very interesting.
Yes. All Portuguese people palatize the S at the coda.
There are two situations where there are more than one pronunciation:
- At the end of a word.
- Followed by a sibilant.
In the standard accent, when an S at the end of a word is followed by word beginning with a vowel, it is pronounced like there were no space.
Thus, um dois três is pronounced "um doish trêsh", but três dois um is pronounced "trêzh dôi zum".
In some accents, it's pronounced like it was followed by a voiced consonant: "trêzh dôi zhum".
In relaxed speech, when an SH or ZH are next to an S or Z, the later are not pronounced, however, some accents do the opposite. Thus, nascer (nash-ser), may be pronounced like na-sher or na-ser.