Just curious, which Wu dialect?
Quite possible, as Wu languages typically have a three way
tenuis-apirated-voiced distinction for some of the consonants.
I have heard unvoiced zh
> in one speaker's pronunciation of Zhao but I have also heard it voiced like in
> judge from Wu speakers.
Wu does not have the retroflex consonants, but it is
understandable that a native Wu speaker might voice some of
them.
> I have even heard it voiced by a resident of
> Shenzhen which is a Mandarin speaking city.
Shenzhen is "Mandarin speaking" in the same sense that Shanghai
is "Mandarin speaking" -- the population has a significant
portion of recent immigrants who are native Mandarin speakers,
while almost all the locals (Yue native speakers) can speak
Mandarin as a second language with varying degree of competence.
And Mandarin is the default language except in private situations
between two locals.
The j in Beijing sounds unvoiced, the zh in Shenzhen is a bit
hard to tell.
>> ----- -----
>>
>> As usual, the c-curl (same as c-cedilla) and z-curl convention
>> causes a lot of confusion as they are used to transcribe both
>> the 'alveolo-palatal' (Mandarin[1]) and 'palatal' (German[2],
>> Malayam[3]) sounds. The Polish sound is described as
>> '(alveolo)-palatal'[4], but the sample sounds 'postalveolar'
>> (i.e., [tS]) or 'palatal' to me. Could it be an allophone
>> situation as in Malayam?
>
> It sounds palatal to me despite being described as alveolo-palatal.
> The Chinese X, however, does sound alveolo-palatal.
How would you advise an Anglophone to pronounce it? I maintain
that "s" is closer than "sh".
> This makes
> their Xi sound odd since I'm used to a palatal adjacent to [i:] and [e:]
> in Malayalam where affricates can be alveolo-palatal only adjacent
> to back vowels and post alveolar only adjacent to a cerebral stop,
> often across a morpheme/ word boundary, or in loan words like
> puncture.
The wikip article on Malayam does not distinguish between
palatal and alveolo-palatal. Is this what you are saying?
front vowel, morpheme initial : postalveolar
back vowel, morpheme initial : postalveolar
front vowel, elsewhere : palatal
back vowel, elsewhere: alveolo-palatal
FWIW the Mandarin alveolo-palatals (x, j, q) appears only
before [i] an [y] (final or medial) and the denti-aveolars
(s, c, z) and the retroflexes (sh, ch, zh) never before
them. The i in <si> etc is not an [i] but a shwa -- the
syllable is a vocalized consonant.
> I find French lignes similarly odd for having an
> alveolo-palatal rather than palatal nasal adjacent to [i:].
It sounds like the regular palatal [j] in the Forvo samples
-- but then I don't have anything to contrast it with.
https://forvo.com/word/ligne/#fr
I suspect we have to bring in the apical-laminal dimension
to clarify all these cases.