Szetc
roger...@mail.com
distinguished only by voice? E.g. sz/zs, fv/vf, etc.
(Intuitively, I'd say no: it just doesn't seem a natural thing to do.)
Miguel Carrasquer
>Does any language use true consonant clusters whose components are
>distinguished only by voice? E.g. sz/zs, fv/vf, etc.
Yes, I'm sure I've seen examples.
Not quite what I had in mind (they're syllabic fricatives
here), but interesting in its own right, is Liangshan Yi
(Nosu)
http://www.und.nodak.edu/dept/linguistics/wp/1997Eatough.PDF
Some words (omitting tone marks):
mmlsz "cloth"
sz "anymore"
sz~ "wood"
fz "ugly" (vs. vz "to buy")
pHz~ "to spit"
=======================
Miguel Carrasquer Vidal
m...@wxs.nl
Miguel Carrasquer
<m...@wxs.nl> wrote:
>On 15 Nov 2005 02:54:21 -0800, roger...@mail.com wrote:
>
>>Does any language use true consonant clusters whose components are
>>distinguished only by voice? E.g. sz/zs, fv/vf, etc.
>
>Yes, I'm sure I've seen examples.
>
>Not quite what I had in mind (they're syllabic fricatives
>here), but interesting in its own right, is Liangshan Yi
>(Nosu)
What I had in mind was Ladefoged & Maddieson, p. 80:
"Sounds that have been described [] as voiced ejectives are,
in our opinion, misnamed. Voiced ejectives have been
reported as contrastive in Zhu|'hõasi []. But as Snyman
makes clear [] these 'voiced ejectives' are prevoiced; the
release is voiceless and from a phonetic point of view they
are clusters of the form [dt']. Clusters involving
obstruents with mixed voicing are very rare in the world's
languages, but they occur in !Xu~ languages and in Kelabit
[]. We have already noted Zhu|'hõasi stops with mixed
voicing in table 3.7 (bpHe "to spit out", dtHa "blanket",
gkHaro "bed"). Further examples [] are given in table 3.17
(dcHau "woman", dc^Hii "to carry straddled on shoulder",
dc'oo "hartebeest", dc^'i "to be wet").
Brian M. Scott
<news:trukn1tmmv0ck0lll...@4ax.com> in
sci.lang:
[...]
> Not quite what I had in mind (they're syllabic fricatives
> here), but interesting in its own right, is Liangshan Yi
> (Nosu)
> http://www.und.nodak.edu/dept/linguistics/wp/1997Eatough.PDF
> Some words (omitting tone marks):
> mmlsz "cloth"
Where (if I'm reading it correctly) the first [m] is
voiceless, and the [ml] is ligatured to represent
simultaneous articulation. This isn't nearly so bad as some
of the N. Amer. languages, but it's still a bit on the
baroque side. (I do like footnote 3: 'Xie Zhili (p.c.)
reports that in certain areas (where the northwestern and
southwestern varieties of Lianghshan Yi come together) there
is bilabial trilling that accompanies the palatalized
postalveolar onsets. When I heard him demonstrate, I
thought of Donald Duck.')
[...]
Brian
Ruud Harmsen
>Does any language use true consonant clusters whose components are
>distinguished only by voice? E.g. sz/zs, fv/vf, etc.
European Portuguese, when spoken at normal (that is, full) speed, with
all reduceable vowels reduced, is full of them.
http://rudhar.com/foneport/en/not2port.htm#Note16b-Syllables
--
Ruud Harmsen - http://rudhar.com
perromaldido
Polish, as in Sczeczin (Stettin)
Perromaldido
perromaldido
Polish, as in Sczeczin (Stettin)
Perromaldido
Des Small
Right. Now there are more Google hits for <Sczeczin> than for
<Szczeczin>, although all the links actually use the second spelling.
(European low-cost airline Ryanair has started flying there, so the
name is slightly familiar to me.)
Also, I was under the impression that Polish <sz> was [S] and <cz> was
<tS> (i.e., that Polish <z> was slightly similar to Engleesh <s>), and
FWIW Wikipedia thinks so too
(http://en.wikipedia.org/wiki/Polish_language#Orthography).
So what has <Szczeczin> got to do with the OP's question?
Des
wants to go to Woodge first
Ruud Harmsen
sci.lang:
>Polish, as in Sczeczin (Stettin)
Isn't <cz> just one phoneme? And <szcz> two, but perhaps also just
one? (Russian has a single cyrillic character for the (historically?)
similar sound).
Harlan Messinger
How do you get *one* syllable out of "enigmático"?!
For "Portugal" and "telefonou", wouldn't /R/ and /l/, respectively,
serve as the "vowel"? I'd have to hear them the way you're describing,
but I can't imagine how to reduce these words beyond /pRt 'gaL/ and /tLf
'no/.
Harlan Messinger
/StS/, a fricative followed by the corresponding affricate, isn't
related to the original question about sequences of corresponding voiced
and unvoiced sounds.
Ekkehard Dengler
Harlan Messinger schrieb:
> How do you get *one* syllable out of "enigmático"?!
>
> For "Portugal" and "telefonou", wouldn't /R/ and /l/, respectively,
> serve as the "vowel"? I'd have to hear them the way you're describing,
> but I can't imagine how to reduce these words beyond /pRt 'gaL/ and /tLf
> 'no/.
The exact processes are hard to describe, but are basically a combination of
elision and coarticulation. Unstressed /e/ is routinely elided while
unstressed /u/ is devoiced and shortened to the point where it's effectively
a rounded consonant.
Suppose you want to reduce "por exemplo" to one syllable. Begin with rounded
[p]. Next, (try to) pronounce [r] and [z] simultaneously; when in doubt,
skip the [r]. Follow through with [e~] and end with [p], released into
voiceless rounded [l]. It can be done.
Regards,
Ekkehard
António Marques
Syllable collapse is indeed possible, but not to the extent Ruud
mentions, I think.
For _telefonou_, for instance, you can make the previous syllable end in
/tl/, place the fricative somewhere in the middle, and make [now] the
only syllable actually attributable to the word.
For _enigmatico_ you're in trouble because the e- usually has to be [e]
or [i]. You can get it to [I], but then it's hard to drop the [I]
itself, unlike what's usual. Of course you can do all these things, but
then you risk a break in communication.
Some contractions are more usual than others, and the most extreme cases
are seldom found. Yet it's possible, unlike in some other languages.
Though I doubt that brazilians find spanish easier than portuguese
because of it. Regardless of american myth, spanish phonetics is a whole
world in itself.
--
am
laurus : rhodophyta : brezoneg : smalltalk : stargate
Brian M. Scott
<hmessinger...@comcast.net> wrote in
<news:3u1873F...@individual.net> in sci.lang:
[...]
> How do you get *one* syllable out of "enigmático"?!
Something like [Nma{tk}], with braces indicating
co-articulation? Or perhaps [gma{tk}]?
[...]
Brian
Paul J Kriha
Harlan Messinger <hmessinger...@comcast.net> wrote in message
news:3u18eaF...@individual.net...
> perromaldido wrote:
> > Ruud Harmsen wrote:
> >
> >>On 15 Nov 2005 02:54:21 -0800, roger...@mail.com wrote:
> >>
> >>>Does any language use true consonant clusters whose components are
> >>>distinguished only by voice? E.g. sz/zs, fv/vf, etc.
> >>
> >>European Portuguese, when spoken at normal (that is, full) speed, with
> >>all reduceable vowels reduced, is full of them.
> >>http://rudhar.com/foneport/en/not2port.htm#Note16b-Syllables
> >>
> >>--
> >>Ruud Harmsen - http://rudhar.com
> >
> >
> > Polish, as in Sczeczin (Stettin)
Oh, poppycock, there is no such place anywhere in Poland.
> /StS/, a fricative followed by the corresponding affricate, isn't
> related to the original question about sequences of corresponding voiced
> and unvoiced sounds.
Exactly, that's why in Polish the name is spelled Szczecin as one would expect.
(Even by some of us like me who don't speak Polish :-).
pjk
Ruud Harmsen
Wed, 16 Nov 2005 12:22:42 -0500: Harlan Messinger
<hmessinger...@comcast.net>: in sci.lang:
>How do you get *one* syllable out of "enigmático"?!
All the vowels except á ([a]) are reduced to a colouring of the
consonants.
It may be that this particular example is wrong (native speakers?),
the initial unstressed e is a special case is some respects. This
example is unattested, but many others I heard myself.
>For "Portugal" and "telefonou", wouldn't /R/ and /l/, respectively,
>serve as the "vowel"?
Can, but isn't.
(It's not /R/, but /r/, by the way. Using a uvular sound in this
position is only possible in Brasil).
>I'd have to hear them the way you're describing,
>but I can't imagine how to reduce these words beyond /pRt 'gaL/ and /tLf
>'no/.
I don't understand how they manage either, but they do. There are
links to samples to the (admittedly less spectacular) example "Rádio
Globo Digital", lower in the page.
Ruud Harmsen
>> How do you get *one* syllable out of "enigmático"?!
>> For "Portugal" and "telefonou", wouldn't /R/ and /l/, respectively,
>> serve as the "vowel"? I'd have to hear them the way you're describing,
>> but I can't imagine how to reduce these words beyond /pRt 'gaL/ and /tLf
>> 'no/.
Wed, 16 Nov 2005 23:56:08 +0100: "Ekkehard Dengler"
<ED...@t-online.de>: in sci.lang:
>The exact processes are hard to describe, but are basically a combination of
>elision and coarticulation. Unstressed /e/ is routinely elided while
>unstressed /u/ is devoiced and shortened to the point where it's effectively
>a rounded consonant.
I listened to the satirical TV program "Contra-Informação" yesterday
(no direct URL, they restructured the access, so I'd have to dig in
the Javascript to find it. Cna be accessed via
http://www.rtp.pt/wportal/multimedia/video/index.php though, date 9
Nov 2005).
They speak veeeeerrryyyyy slowly in that show (like bo..a noi..te),
but even then there were those occasional 'rattles' , in stock phrases
like 'todos os Portugueses".
It seems TV Portuguese is becoming slower these days than it used to
be. All the presidential candidates speak very slowly, Cavaco Silva
for example. Or is just that I now understand more than before?
>Suppose you want to reduce "por exemplo" to one syllable.
>Begin with rounded
>[p]. Next, (try to) pronounce [r] and [z] simultaneously; when in doubt,
>skip the [r]. Follow through with [e~] and end with [p], released into
>voiceless rounded [l]. It can be done.
Exactly.
Examples from the above-mentioned satirical show:
2:02: "comigo" (with me)
3:08: "quando percebi" (when I finally understood what was going on)
4:18: "estes árbitros" (these arbiters)
3:53: é que às vezes os defesas do Porto" (the thing is sometimes
Porto's defenders ...)
(The part between 'vez' and 'fes' is completely "rolled into one".)
4:07: "prejudicial" (hazardous, damaging): one syllable, maybe two,
but no more. Not five, as the spelling suggests, and as the
corresponding Spanish, Italian or French word (or Brasilian
Portuguese, for that matter) would have..
Again, this show is in exaggeratedly slow speech. In normal
conversation, nearly everything is in this contracted mode. Yet,
often, it can actually be understood, after a lot of practice ...
Ruud Harmsen
<b.s...@csuohio.edu>: in sci.lang:
>> How do you get *one* syllable out of "enigmático"?!
>
>Something like [Nma{tk}],
No, the [n] and the [g] don't assimilate. There is very little
assimilation between consonsants in Portuguese. Voiced consonants may
get devoiced, but without becoming the same as the corresponing
voiceless consonant. And place of articulation practically always
stays distinct.
>with braces indicating
>co-articulation? Or perhaps [gma{tk}]?
--
Ruud Harmsen
sci.lang:
>For _enigmatico_ you're in trouble because the e- usually has to be [e]
>or [i]. You can get it to [I], but then it's hard to drop the [I]
>itself, unlike what's usual. Of course you can do all these things, but
>then you risk a break in communication.
There are lots of such breaks when I try to understand spoken
Portuguese. (;-)
Now that I'm getting better, I often still have that "huh??" feeling
at the end of a phrase, and then I sort of see the whole thing rolled
back in a mental movie, re-assembled into full forms, and I realise
what it must have been that led to what I actually heard. If the pause
is long enough for this mental process to evolve, I can understand the
meaning, but if the next full speed sentence has already started, I
often "derail", which means switching from understanding nearly
everything to nothing at all.There is nothing in between. It's either
95% or 0%. Formerly, I was more often at zero, now I am at 95% for
longer periods. But only if the speaker adheres to a standard Lisbon
or Coimbra accent. The slighest regional accent can put me off the
track already.
>Some contractions are more usual than others, and the most extreme cases
>are seldom found.
All the time, in my experience. Especially in heated TV debates
between a politician and a tv reporter who tries to make them say the
wrong things.
John Atkinson
"Ruud Harmsen" <realemail...@rudhar.com.invalid> wrote ...
> 16 Nov 2005 08:44:05 -0800: "perromaldido" <kill...@sofcom.com>:
>
>>Polish, as in Sczeczin (Stettin)
>
> Isn't <cz> just one phoneme? And <szcz> two, but perhaps also just
> one? (Russian has a single cyrillic character for the (historically?)
> similar sound).
More to the point, both <sz> and <cz> are voiceless, however many phonemes
their combination forms.
John.
Ekkehard Dengler
Ruud Harmsen schrieb:
> Harlan Messinger schrieb:
> >I'd have to hear them the way you're describing,
> >but I can't imagine how to reduce these words beyond /pRt 'gaL/ and /tLf
> >'no/.
>
> I don't understand how they manage either, but they do. There are
> links to samples to the (admittedly less spectacular) example "Rádio
> Globo Digital", lower in the page.
If you want spectacular examples, I suggest looking out for the very popular
trisyllabic version of "Futebol Clube do Porto".
Regards,
Ekkehard
Ruud Harmsen
<ED...@t-online.de>: in sci.lang:
>If you want spectacular examples, I suggest looking out for the very popular
>trisyllabic version of "Futebol Clube do Porto".
I'll wait until my compatriot Co Adriaanse can properly say that. Will
take a long time. So far he only speaks English, which is duly
translated. Ronald Koeman in Lisbon still always speaks (rather good)
Spanish, which is left untranslated.
Harlan Messinger
>
> For "Portugal" and "telefonou", wouldn't /R/ and /l/, respectively,
> serve as the "vowel"? I'd have to hear them the way you're describing,
> but I can't imagine how to reduce these words beyond /pRt 'gaL/ and /tLf
> 'no/.
Rather, /tLf 'now/?
Harlan Messinger
> Harlan Messinger schrieb:
>
>
>>How do you get *one* syllable out of "enigmático"?!
>>
>>For "Portugal" and "telefonou", wouldn't /R/ and /l/, respectively,
>>serve as the "vowel"? I'd have to hear them the way you're describing,
>>but I can't imagine how to reduce these words beyond /pRt 'gaL/ and /tLf
>>'no/.
>
>
> The exact processes are hard to describe, but are basically a combination of
> elision and coarticulation. Unstressed /e/ is routinely elided while
> unstressed /u/ is devoiced and shortened to the point where it's effectively
> a rounded consonant.
>
> Suppose you want to reduce "por exemplo" to one syllable. Begin with rounded
> [p]. Next, (try to) pronounce [r] and [z] simultaneously; when in doubt,
> skip the [r].
Is that anything like the r-colored /d/ and /n/ sounds in Swedish? (And
what's the proper term for those?) That would help me slightly to
understand the articulation of "Portugal".
Miguel Carrasquer
<hmessinger...@comcast.net> wrote:
>Ekkehard Dengler wrote:
>> Suppose you want to reduce "por exemplo" to one syllable. Begin with rounded
>> [p]. Next, (try to) pronounce [r] and [z] simultaneously; when in doubt,
>> skip the [r].
>
>Is that anything like the r-colored /d/ and /n/ sounds in Swedish?
No.
>(And what's the proper term for those?)
Retroflex.
Ruud Harmsen
<hmessinger...@comcast.net>: in sci.lang:
>> Suppose you want to reduce "por exemplo" to one syllable. Begin with rounded
>> [p]. Next, (try to) pronounce [r] and [z] simultaneously; when in doubt,
>> skip the [r].
>
>Is that anything like the r-colored /d/ and /n/ sounds in Swedish?
Retroflex r + alveolar (dental?) t or d ==> retroflex t or d.
Ekkehard Dengler
Harlan Messinger schrieb:
Only if you want to sound like a northerner.
Regards,
Ekkehard
António Marques
>>> For "Portugal" and "telefonou", wouldn't /R/ and /l/, respectively,
>>> serve as the "vowel"? I'd have to hear them the way you're describing,
>>> but I can't imagine how to reduce these words beyond /pRt 'gaL/ and /tLf
>>> 'no/.
>>
>> Rather, /tLf 'now/?
>
> Only if you want to sound like a northerner.
Do you know anyone for whom toro and touro are the same? The greatest
reduction of /ou/ I've seen is [o:] (where /o/ would be [o] - add to
that that <ou> remains unchanged in unstressed position).
Really, claiming such reductions are natural is a bit like saying you
can cram 'er hat es mir nicht gesagt' into 3 syllables (which I think
can even sound ok, /rhatsm rnICt kzakt/ or something).
António Marques
>> If you want spectacular examples, I suggest looking out for the very popular
>> trisyllabic version of "Futebol Clube do Porto".
>
> I'll wait until my compatriot Co Adriaanse can properly say that. Will
> take a long time.
Maybe forever. I can't. I think you have to have been born there and
have been a registered supporter before you've even had a birth certificate.
> So far he only speaks English, which is duly
> translated. Ronald Koeman in Lisbon still always speaks (rather good)
> Spanish, which is left untranslated.
One can understand him much better than Camacho.
All in all, however, the only example of remote relevance for the
original question may be 'Fevereiro' /fv/?, with great licence.
Ekkehard Dengler
António Marques schrieb:
> Ekkehard Dengler wrote:
>
> >>> For "Portugal" and "telefonou", wouldn't /R/ and /l/, respectively,
> >>> serve as the "vowel"? I'd have to hear them the way you're describing,
> >>> but I can't imagine how to reduce these words beyond /pRt 'gaL/ and
/tLf
> >>> 'no/.
> >>
> >> Rather, /tLf 'now/?
> >
> > Only if you want to sound like a northerner.
>
> Do you know anyone for whom toro and touro are the same?
No, I don't; "toro" is ['tO:ru] and "touro" is ['to:ru], except in the
north, where it's typically pronounced with a diphthong. How's that
relevant, though? I don't think I quite understand what you're implying
here. Anyway, the fact remains that there's no such thing as an /ou/ phoneme
in either the standard pronunciation or, for that matter, any regional
variety spoken outside the north.
> The greatest
> reduction of /ou/ I've seen is [o:] (where /o/ would be [o] - add to
> that that <ou> remains unchanged in unstressed position).
Straightforward [o:] is a perfectly unremarkable realisation of the /o/
phoneme, not some kind of unusual reduction.
> Really, claiming such reductions are natural is a bit like saying you
> can cram 'er hat es mir nicht gesagt' into 3 syllables (which I think
> can even sound ok, /rhatsm rnICt kzakt/ or something).
Except that you'll never hear anything like that from a German newsreader
while José Rodrigues dos Santos routinely manages to cram "Futebol Clube do
Porto" into three syllables -- and gets away with it. He did it again
tonight.
Regards,
Ekkehard
Ekkehard Dengler
António Marques schrieb:
> All in all, however, the only example of remote relevance for the
> original question may be 'Fevereiro' /fv/?, with great licence.
Do Portuguese place-names count as well? If so: "Sesimbra" [sz].
Regards,
Ekkehard
António Marques
>> Do you know anyone for whom toro and touro are the same?
>
> No, I don't; "toro" is ['tO:ru]
Excpet when it's [toru].
> Anyway, the fact remains that there's no such thing as an /ou/ phoneme
> in either the standard pronunciation or, for that matter, any regional
> variety spoken outside the north.
There is no /ou/ phoneme. There's /o/ followed by /u/, which is realised
distinctively as varieties of [ow] in the north, and less distinctively
so tending towards [o] as you walk southwards. However, I don't think it
come to merge with plain /o/ in any part of the territory.
A parallel phenomenon happens with <ei>, though the gradients are
different. I don't think 'pe^ra' and 'beira' merge anywhere.
>> The greatest
>> reduction of /ou/ I've seen is [o:] (where /o/ would be [o] - add to
>> that that <ou> remains unchanged in unstressed position).
>
> Straightforward [o:] is a perfectly unremarkable realisation of the /o/
> phoneme, not some kind of unusual reduction.
But unstressed /ou/ is longer than stressed /o/, and even if it were as
long, it would be the only portuguese phoneme that wouldn't be shorter
when unstressed, which isn't a very useful analysis, and what do you do
with de facto unstressed short [o], as in co-piloto, cofragem, etc? What
phoneme is it? Those words don't sound like *cou-piloto or *coufragem.
(NB 'Coutinho' and 'couto' have simple [o])
For another stressed example, the vowel in manjedoura is different from
the one in transportadora.
Of course with extreme reduction, everything may get reduced. Even d's
disappear, as I pointed out a while ago. But that doesn't mean they
don't exist.
>> Really, claiming such reductions are natural is a bit like saying you
>> can cram 'er hat es mir nicht gesagt' into 3 syllables (which I think
>> can even sound ok, /rhatsm rnICt kzakt/ or something).
>
> Except that you'll never hear anything like that from a German newsreader
> while José Rodrigues dos Santos routinely manages to cram "Futebol Clube do
> Porto" into three syllables -- and gets away with it. He did it again
> tonight.
Are you sure you won't hear anything like that from a german newsreader?
Because I can assure you that when I hear german on the TV they do
manage to do stuff of that sort (except on Deutsche Welle).
António Marques
>> All in all, however, the only example of remote relevance for the
>> original question may be 'Fevereiro' /fv/?, with great licence.
>
> Do Portuguese place-names count as well? If so: "Sesimbra" [sz].
Yes - I looked for one with [SZ] or [ZS], but couldn't find any. I think
it's harder to find a voiced-unvoiced sequence, as it ould imply a very
audible vowel between. OTOH, there's plenty of [sZ] or [Zs] or even
[zZ], [Ss], etc ([Z] is less voiced near [s] but different from [S]).
Ekkehard Dengler
> Ekkehard Dengler wrote:
>
> >> Do you know anyone for whom toro and touro are the same?
> >
> > No, I don't; "toro" is ['tO:ru]
>
> Excpet when it's [toru].
I don't think I've ever heard "toro" with /o/, and in any case I doubt that
the stressed vowel would be short.
> > Anyway, the fact remains that there's no such thing as an /ou/ phoneme
> > in either the standard pronunciation or, for that matter, any regional
> > variety spoken outside the north.
>
> There is no /ou/ phoneme. There's /o/ followed by /u/, which is realised
> distinctively as varieties of [ow] in the north, and less distinctively
> so tending towards [o] as you walk southwards. However, I don't think it
> come to merge with plain /o/ in any part of the territory.
The two have certainly merged in standard Portuguese.
> >> The greatest
> >> reduction of /ou/ I've seen is [o:] (where /o/ would be [o] - add to
> >> that that <ou> remains unchanged in unstressed position).
> >
> For another stressed example, the vowel in manjedoura is different from
> the one in transportadora.
No, it isn't, for most speakers. In both standard Portuguese and many
regional varieties, "manjedoura" rhymes with "transportadora".
Regards,
Ekkehard
Ruud Harmsen
sci.lang:
>Yes - I looked for one with [SZ] or [ZS], but couldn't find any.
Jesus [ZzuS]. Doesn't quite count, not the same place of articulation.
Ruud Harmsen
sci.lang:
>Ekkehard Dengler wrote:
>
>>> Do you know anyone for whom toro and touro are the same?
>>
>> No, I don't; "toro" is ['tO:ru]
>
>Excpet when it's [toru].
Which is when?
>Of course with extreme reduction, everything may get reduced. Even d's
>disappear, as I pointed out a while ago. But that doesn't mean they
>don't exist.
In my experience, all the consonants are always still there, no matter
how strong the reduction. A voiced lax [d] can become unvoiced lax
[d-with-circle-underneath] (as is "o descobridor"), but that's still
different from voiceless tense [t].
>> Except that you'll never hear anything like that from a German newsreader
>> while José Rodrigues dos Santos routinely manages to cram "Futebol Clube do
>> Porto" into three syllables -- and gets away with it. He did it again
>> tonight.
>
>Are you sure you won't hear anything like that from a german newsreader?
Seem rather unlikely.
>Because I can assure you that when I hear german on the TV they do
>manage to do stuff of that sort (except on Deutsche Welle).
When I listen to a Hungarian radio station I hear all the vowels and
all the syllables, without being able to understand anything, simply
because I don't know the language.
In Portuguese, I either stay on track, or I get derailed, lose sync.
European Portuguese can only be heard if you understand it first. What
you don't understand (as a foreign learner) you also don't hear, so
you can't look it up in a dictionary. In other languages this is
possible. EP is really special in this respect.
Ekkehard Dengler
António Marques schrieb:
> Ekkehard Dengler wrote:
> There is no /ou/ phoneme. There's /o/ followed by /u/, which is realised
> distinctively as varieties of [ow] in the north, and less distinctively
> so tending towards [o] as you walk southwards. However, I don't think it
> come to merge with plain /o/ in any part of the territory.
>
> A parallel phenomenon happens with <ei>, though the gradients are
> different. I don't think 'pêra' and 'beira' merge anywhere.
They've definitely merged in parts of the Alentejo. You must have heard the
saying "Os alentejanos tiram o <i> do lêti e põem no caféi." It's basically
true.
> But unstressed /ou/ is longer than stressed /o/,
In most of the country, it clearly isn't: "estouvado" [Sto.'va:du], "estojo"
['Sto:Zu].
> >> Really, claiming such reductions are natural is a bit like saying you
> >> can cram 'er hat es mir nicht gesagt' into 3 syllables (which I think
> >> can even sound ok, /rhatsm rnICt kzakt/ or something).
> >
> > Except that you'll never hear anything like that from a German
newsreader
> > while José Rodrigues dos Santos routinely manages to cram "Futebol Clube
do
> > Porto" into three syllables -- and gets away with it. He did it again
> > tonight.
>
> Are you sure you won't hear anything like that from a german newsreader?
Yes, very. Except in private.
Regards,
Ekkehard
António Marques
>>>> Do you know anyone for whom toro and touro are the same?
>>>
>>> No, I don't; "toro" is ['tO:ru]
>>
>> Excpet when it's [toru].
>
> Which is when?
Every time, in my environment at least, which isn't northerner. Plural
[tOruS], though.
>> Of course with extreme reduction, everything may get reduced. Even
>> d's disappear, as I pointed out a while ago. But that doesn't mean
>> they don't exist.
>
> In my experience, all the consonants are always still there, no
> matter how strong the reduction. A voiced lax [d] can become unvoiced
> lax [d-with-circle-underneath] (as is "o descobridor"), but that's
> still different from voiceless tense [t].
Though I really don't know what happened to the d in 'os defesas do
Porto' aka 'ujefesajuport'.
António Marques
>> Yes - I looked for one with [SZ] or [ZS], but couldn't find any.
>
> Jesus [ZzuS]. Doesn't quite count, not the same place of
> articulation.
'Gestacao' [ZSt@s'@~w] when spoken quickly.
António Marques
>> Yes - I looked for one with [SZ] or [ZS], but couldn't find any.
>
> Jesus [ZzuS]. Doesn't quite count, not the same place of
> articulation.
'Gestacao' [ZSt@s'@~w] when spoken quickly.
António Marques
>> Yes - I looked for one with [SZ] or [ZS], but couldn't find any.
>
> Jesus [ZzuS]. Doesn't quite count, not the same place of
> articulation.
'Gestacao' [ZSt@s'@~w] when spoken quickly.
António Marques
>>>> Do you know anyone for whom toro and touro are the same?
>>> No, I don't; "toro" is ['tO:ru]
>> Excpet when it's [toru].
>
> I don't think I've ever heard "toro" with /o/,
It's how I've always heard it.
> and in any case I doubt that the stressed vowel would be short.
It isn't very long, but I put no length mark there because I think it's
superfluous.
>>> Anyway, the fact remains that there's no such thing as an /ou/
>>> phoneme in either the standard pronunciation or, for that matter,
>>> any regional variety spoken outside the north.
>>
>> There is no /ou/ phoneme. There's /o/ followed by /u/, which is
>> realised distinctively as varieties of [ow] in the north, and less
>> distinctively so tending towards [o] as you walk southwards.
>> However, I don't think it come to merge with plain /o/ in any part
>> of the territory.
>
> The two have certainly merged in standard Portuguese.
No, they have not. Check for misspellings - they're hard to find.
Just because <ou> isn't [ow] south of, say, S. Pedro do Sul (which, for
the uninitiated, is to the NW of Viseu), that doesn't mean it merges
with <o^>. It's always a tad bit more rounded, or longer, or more
closed, or the 3. I admit it may be difficult to hear, but that's the
way it is, it's not may fault.
>> A parallel phenomenon happens with <ei>, though the gradients are
>> different. I don't think 'pêra' and 'beira' merge anywhere.
>
> They've definitely merged in parts of the Alentejo. You must have
> heard the saying "Os alentejanos tiram o <i> do lêti e põem no
> caféi." It's basically true.
Here we may pick up another issue, that of the speakers' preception of
their pronunciation, for one, and of that of others. For all the whole
country cares, Viseu* people substitute [S] for [s] and [Z] for [z]. But
you know that's not correct.
(*) Let alone that laminal sibilants affect the entire NE, not Viseu in
particular.
Maybe there are parts of the Alentejo where they merge; but the usual
phenomenon is
[eji] (very short [i]) ->
[ej] ->
[ej] (with nearly imperceptible glide) ->
something like [ee] ->
tremolo [e:] or something
whereas /e/, even when [e:], doesn't have such a marked tremolo effect.
I admit it may have in places, or that <ei> may lose it, but it's no
simple matter of entering the Alentejo, hearing some kind of [e] for
<ei>, and deciding that <ei> is plain /e/.
Is there an [a] in german dialectal 'wiad' or 'woad'?
>> But unstressed /ou/ is longer than stressed /o/,
>
> In most of the country, it clearly isn't: "estouvado" [Sto.'va:du],
> "estojo" ['Sto:Zu].
_Estojo_ is special by virtue of the /Z/ - it goes all the way from
'estoujo' to 'estoijo'. 'Estofo' is more to the point, and hasn't a
longer vowel than 'estouvado'.
>>>> The greatest reduction of /ou/ I've seen is [o:] (where /o/
>>>> would be [o] - add to that that <ou> remains unchanged in
>>>> unstressed position).
>>
>> For another stressed example, the vowel in manjedoura is different
>> from the one in transportadora.
>
> No, it isn't, for most speakers. In both standard Portuguese and many
> regional varieties, "manjedoura" rhymes with "transportadora".
-our[ao] and -or[ao] words are kept distinct.
Ekkehard Dengler
António Marques schrieb:
> Ekkehard Dengler wrote:
>
> >>>> Do you know anyone for whom toro and touro are the same?
> >>> No, I don't; "toro" is ['tO:ru]
> >> Excpet when it's [toru].
> >
> > I don't think I've ever heard "toro" with /o/,
>
> It's how I've always heard it.
I'm surprised, but then I'm not really familiar with northern accents.
> > and in any case I doubt that the stressed vowel would be short.
>
> It isn't very long, but I put no length mark there because I think it's
> superfluous.
Not if you're using square brackets and talking about vowel length.
> >>> Anyway, the fact remains that there's no such thing as an /ou/
> >>> phoneme in either the standard pronunciation or, for that matter,
> >>> any regional variety spoken outside the north.
> >>
> >> There is no /ou/ phoneme. There's /o/ followed by /u/, which is
> >> realised distinctively as varieties of [ow] in the north, and less
> >> distinctively so tending towards [o] as you walk southwards.
> >> However, I don't think it come to merge with plain /o/ in any part
> >> of the territory.
> >
> > The two have certainly merged in standard Portuguese.
>
> No, they have not. Check for misspellings - they're hard to find.
Not at all. I know people who spell "estendedouro" with "or", for example.
Try giving dictation to schoolchildren from the south. You'll be surprised.
> Just because <ou> isn't [ow] south of, say, S. Pedro do Sul (which, for
> the uninitiated, is to the NW of Viseu), that doesn't mean it merges
> with <o^>. It's always a tad bit more rounded, or longer, or more
> closed, or the 3. I admit it may be difficult to hear, but that's the
> way it is, it's not may fault.
No, you're mistaken. I'm afraid you're imagining things. For millions of
speakers, <ou> is /o/.
> >> A parallel phenomenon happens with <ei>, though the gradients are
> >> different. I don't think 'pêra' and 'beira' merge anywhere.
> >
> > They've definitely merged in parts of the Alentejo. You must have
> > heard the saying "Os alentejanos tiram o <i> do lêti e põem no
> > caféi." It's basically true.
> whereas /e/, even when [e:], doesn't have such a marked tremolo effect.
> I admit it may have in places, or that <ei> may lose it, but it's no
> simple matter of entering the Alentejo, hearing some kind of [e] for
> <ei>, and deciding that <ei> is plain /e/.
> Is there an [a] in german dialectal 'wiad' or 'woad'?
Sorry?
> >>>> The greatest reduction of /ou/ I've seen is [o:] (where /o/
> >>>> would be [o] - add to that that <ou> remains unchanged in
> >>>> unstressed position).
> >>
> >> For another stressed example, the vowel in manjedoura is different
> >> from the one in transportadora.
> >
> > No, it isn't, for most speakers. In both standard Portuguese and many
> > regional varieties, "manjedoura" rhymes with "transportadora".
>
> -our[ao] and -or[ao] words are kept distinct.
António, I hate to say it, but you're simply wrong.
Regards,
Ekkehard
António Marques
>> Yes - I looked for one with [SZ] or [ZS], but couldn't find any.
>
> Jesus [ZzuS]. Doesn't quite count, not the same place of
> articulation.
'Gestacao' [ZSt@s'@~w] when spoken quickly.
António Marques
>>>> Do you know anyone for whom toro and touro are the same?
>>> No, I don't; "toro" is ['tO:ru]
>> Excpet when it's [toru].
>
> I don't think I've ever heard "toro" with /o/,
It's how I've always heard it.
> and in any case I doubt that the stressed vowel would be short.
It isn't very long, but I put no length mark there because I think it's
superfluous.
>>> Anyway, the fact remains that there's no such thing as an /ou/
>>> phoneme in either the standard pronunciation or, for that matter,
>>> any regional variety spoken outside the north.
>>
>> There is no /ou/ phoneme. There's /o/ followed by /u/, which is
>> realised distinctively as varieties of [ow] in the north, and less
>> distinctively so tending towards [o] as you walk southwards.
>> However, I don't think it come to merge with plain /o/ in any part
>> of the territory.
>
> The two have certainly merged in standard Portuguese.
No, they have not. Check for misspellings - they're hard to find.
Just because <ou> isn't [ow] south of, say, S. Pedro do Sul (which, for
the uninitiated, is to the NW of Viseu), that doesn't mean it merges
with <o^>. It's always a tad bit more rounded, or longer, or more
closed, or the 3. I admit it may be difficult to hear, but that's the
way it is, it's not may fault.
>> A parallel phenomenon happens with <ei>, though the gradients are
>> different. I don't think 'pêra' and 'beira' merge anywhere.
>
> They've definitely merged in parts of the Alentejo. You must have
> heard the saying "Os alentejanos tiram o <i> do lêti e põem no
> caféi." It's basically true.
Here we may pick up another issue, that of the speakers' preception of
their pronunciation, for one, and of that of others. For all the whole
country cares, Viseu* people substitute [S] for [s] and [Z] for [z]. But
you know that's not correct.
(*) Let alone that laminal sibilants affect the entire NE, not Viseu in
particular.
Maybe there are parts of the Alentejo where they merge; but the usual
phenomenon is
[eji] (very short [i]) ->
[ej] ->
[ej] (with nearly imperceptible glide) ->
something like [ee] ->
tremolo [e:] or something
whereas /e/, even when [e:], doesn't have such a marked tremolo effect.
I admit it may have in places, or that <ei> may lose it, but it's no
simple matter of entering the Alentejo, hearing some kind of [e] for
<ei>, and deciding that <ei> is plain /e/.
Is there an [a] in german dialectal 'wiad' or 'woad'?
>> But unstressed /ou/ is longer than stressed /o/,
>
> In most of the country, it clearly isn't: "estouvado" [Sto.'va:du],
> "estojo" ['Sto:Zu].
_Estojo_ is special by virtue of the /Z/ - it goes all the way from
'estoujo' to 'estoijo'. 'Estofo' is more to the point, and hasn't a
longer vowel than 'estouvado'.
>>>> The greatest reduction of /ou/ I've seen is [o:] (where /o/
>>>> would be [o] - add to that that <ou> remains unchanged in
>>>> unstressed position).
>>
>> For another stressed example, the vowel in manjedoura is different
>> from the one in transportadora.
>
> No, it isn't, for most speakers. In both standard Portuguese and many
> regional varieties, "manjedoura" rhymes with "transportadora".
-our[ao] and -or[ao] words are kept distinct.
Ruud Harmsen
sci.lang:
>Just because <ou> isn't [ow] south of, say, S. Pedro do Sul (which, for
>the uninitiated, is to the NW of Viseu), that doesn't mean it merges
>with <o^>. It's always a tad bit more rounded, or longer, or more
>closed, or the 3. I admit it may be difficult to hear, but that's the
>way it is, it's not may fault.
I always had the feeling there was some difference too, without being
able to say what. I don't know how much of that is due to expectation
from the spelling. I knew the spelling long before I read about (=
didn't myself discover by listening) the phonetic similarity or
identity of ou and ô in Portuguese.
>(*) Let alone that laminal sibilants affect the entire NE, not Viseu in
>particular.
Meaning there is a difference between paço and passo? Vianna mentioned
that in 1892, for Camões' time and today's (his today, that is)
Trás-os-Montes. http://purl.pt/146, pages 90-93.
===
70. Parece averiguado que há três séculus a pronuncia [sic, não
"pronúncia"; RH] do português de Lisbôa differia da actual nos
sequintes pontos, que deveriam ser tidos em attenção numa leitura
rigorosa do poêma.
I. O s inicial e ss mediaes (saber, passo) differençavam-se do ç ou c
antes de e, i, em que, como ainda hoje em Trás-os-Montes e parte do
Minho e Beiras, eram preferidos com a superficie anterior do ápice da
lingua, aproximando esse ápice, assim cóncavo, das gengivas dos
incisivos superiores, posição que denominámos reversa, o que indicámos
pelo sýmbolo s [com ponto debaixo] isto é, [s longo]. Êste valor do s
mantinha-se-lhe depois de consoante, quando final na pausa, e antes
das consoantes surdas p, t, c, qu, ç, f, x. Portanto o vocábulo p a s
s o era differente de p a ç o, êstes pronunciava-se êstes [não consio
lê-lo exactamente] (V.42).
II. Semelhantemente, z inicial ou medial differençava-se de s sonoro
entre vogaes, em que êste era, como é em Trás-os-Montes e parte do
Minho, um z proferido com os órgãos a mesma posição que fica descrita
para o s inicial (V. p. 47), distinguindo-se conseguintemente o
vocábulo c o s e r ([?????]) do vocábulo c o z e r.
Esta pronuncia do s pode designar-se por s[com ponto debaixo] ou z[com
ponto debaixo]. Êste som tinha igualmente o s antes de consoante
sonora, b, d, g, gu, z, j, v, m, n, nh, r, l, lh, quer dentro de um
vocábulo, quer de um para outro vocábulo, e do mesmo modo o s final na
junção com a vogal inicial sequinte : azarmas [diacritos fonéticas
difícil a ler; RH], e não azarmas = as armas.
O z final, porém, assim como a z interno, muito raro, antes de
consoante surda, proferia-se, o que acontece actualmente em
Trás-os-Montes, como ç; assim paz, luz pállida, luz azul, luz verde,
pronunciavam-se respectivamente: paç, luçpálida, luzazúl, luz(z)vêrde.
===
Ruud Harmsen
sci.lang:
>Is there an [a] in german dialectal 'wiad' or 'woad'?
(For <wird>?) My German-Dutch Van Dale dictionary write turned a
(Sampa [6]) for all final <r>'s.
Ruud Harmsen
sci.lang:
>Though I really don't know what happened to the d in 'os defesas do
>Porto' aka 'ujefesajuport'.
I clearly hear in the (admittedly rather slow) sample I posted earlier
(Contra-informação, 3:53,
http://www.rtp.pt/wportal/multimedia/video/index.php
António Marques
>>>>>> Do you know anyone for whom toro and touro are the same?
>>>>> No, I don't; "toro" is ['tO:ru]
>>>> Excpet when it's [toru].
>>> I don't think I've ever heard "toro" with /o/,
>> It's how I've always heard it.
>
> I'm surprised, but then I'm not really familiar with northern accents.
It's not northern.
>>> and in any case I doubt that the stressed vowel would be short.
>> It isn't very long, but I put no length mark there because I think it's
>> superfluous.
>
> Not if you're using square brackets and talking about vowel length.
No, square brackets indicate pronunciation. Just how much detail should
be included depends on various factors. In the case I was presenting,
Stressed Unstressed
<ou> [o:.] [o:]
<o> [o.] [o]
>>>>> Anyway, the fact remains that there's no such thing as an /ou/
>>>>> phoneme in either the standard pronunciation or, for that matter,
>>>>> any regional variety spoken outside the north.
>>>> There is no /ou/ phoneme. There's /o/ followed by /u/, which is
>>>> realised distinctively as varieties of [ow] in the north, and less
>>>> distinctively so tending towards [o] as you walk southwards.
>>>> However, I don't think it come to merge with plain /o/ in any part
>>>> of the territory.
>>> The two have certainly merged in standard Portuguese.
>> No, they have not. Check for misspellings - they're hard to find.
>
> Not at all. I know people who spell "estendedouro" with "or", for example.
> Try giving dictation to schoolchildren from the south. You'll be surprised.
Hey, I know schoolchildren from the south (does Setubal count?). There
is stuff that is regularly misspelt in portuguese, but <ou> isn't one,
except in some words where the spelling doesn't reflect real phonology.
In this case it's 'estendor' or something.
Look for misspellings.
>> Just because <ou> isn't [ow] south of, say, S. Pedro do Sul (which, for
>> the uninitiated, is to the NW of Viseu), that doesn't mean it merges
>> with <o^>. It's always a tad bit more rounded, or longer, or more
>> closed, or the 3. I admit it may be difficult to hear, but that's the
>> way it is, it's not may fault.
>
> No, you're mistaken. I'm afraid you're imagining things. For millions of
> speakers, <ou> is /o/.
Just because you can't hear it, it doesn't mean the difference doesn't
exist.
>> Is there an [a] in german dialectal 'wiad' or 'woad'?
>
> Sorry?
When peple spell their dialect with 'wiad' for 'wird', you might take
that as evidence that their wird has some kind of [a]. Just as you might
assume that if someone says they say 'lo^c,a' for 'louc,a', their word
is evidence that they have plain /o/ there; of course, if they then
proceed to say it's a longer o^ than the usual one, you can discount
that as spelling induced imagination and their word is no longer good.
>>>>>> The greatest reduction of /ou/ I've seen is [o:] (where /o/
>>>>>> would be [o] - add to that that <ou> remains unchanged in
>>>>>> unstressed position).
>>>> For another stressed example, the vowel in manjedoura is different
>>>> from the one in transportadora.
>>> No, it isn't, for most speakers. In both standard Portuguese and many
>>> regional varieties, "manjedoura" rhymes with "transportadora".
>> -our[ao] and -or[ao] words are kept distinct.
>
> António, I hate to say it, but you're simply wrong.
It's really no big deal - you've decided that <ou> is /o/ and nothing
can make you change your mind. I don't have a problem with that, all I
care is to present the information I know so that others can be... more
informed.
António Marques
>> Is there an [a] in german dialectal 'wiad' or 'woad'?
>
> (For <wird>?) My German-Dutch Van Dale dictionary write turned a
> (Sampa [6]) for all final <r>'s.
Is that the same you use for portuguese /@/?
From a distance, I find german dialects so similar between themselves
and to the standard, that I have trouble understanding why such a fuss
is made about their differences. Isn't english as dialectally diverse,
if we consider phonetics only?
Ekkehard Dengler
António Marques schrieb:
> Ekkehard Dengler wrote:
> >> Just because <ou> isn't [ow] south of, say, S. Pedro do Sul (which, for
> >> the uninitiated, is to the NW of Viseu), that doesn't mean it merges
> >> with <o^>. It's always a tad bit more rounded, or longer, or more
> >> closed, or the 3. I admit it may be difficult to hear, but that's the
> >> way it is, it's not may fault.
> >
> > No, you're mistaken. I'm afraid you're imagining things. For millions of
> > speakers, <ou> is /o/.
>
> Just because you can't hear it, it doesn't mean the difference doesn't
> exist.
Yes, that's exactly what it means, unless by "you" you mean me personally.
> >> Is there an [a] in german dialectal 'wiad' or 'woad'?
> >
> > Sorry?
>
> When peple spell their dialect with 'wiad' for 'wird', you might take
> that as evidence that their wird has some kind of [a]. Just as you might
> assume that if someone says they say 'lo^c,a' for 'louc,a', their word
> is evidence that they have plain /o/ there; of course, if they then
> proceed to say it's a longer o^ than the usual one, you can discount
> that as spelling induced imagination and their word is no longer good.
See below.
> >>>>>> The greatest reduction of /ou/ I've seen is [o:] (where /o/
> >>>>>> would be [o] - add to that that <ou> remains unchanged in
> >>>>>> unstressed position).
> >>>> For another stressed example, the vowel in manjedoura is different
> >>>> from the one in transportadora.
> >>> No, it isn't, for most speakers. In both standard Portuguese and many
> >>> regional varieties, "manjedoura" rhymes with "transportadora".
> >> -our[ao] and -or[ao] words are kept distinct.
> >
> > António, I hate to say it, but you're simply wrong.
>
> It's really no big deal - you've decided that <ou> is /o/ and nothing
> can make you change your mind. I don't have a problem with that, all I
> care is to present the information I know so that others can be... more
> informed.
I appreciate you listen very closely to the way people speak, but I regret
to say I'm getting the impression that you don't really know how to
interpret what you hear.
I can assure you I'm not confusing eye dialect with reality. If you don't
want to take my word for it, let me suggest reading about the phonology of
Portuguese and/or talking to Portuguese linguists about the subject. An
excellent starting point would be Paul Teyssier's "História da Língua
Portuguesa". Here's what he has to say about "ou" and "ei":
Page 52 (7th Portuguese edition, 1997):
<O ditongo "ou", isto é, [ow], passou a [o] no actual português comum [...].
Esta monotongação começou provavelmente a manifestar-se no século XVII.
Invadiu todo o Sul e a maior parte do Centro de Portugal, mas no resto do
país, ou seja, uma vez mais, no Norte, o antigo ditongo "ou" [ou] continua
vivo. O limite do fenómeno parte do oeste da região de Aveiro e forma uma
espécie de S que alcança a fronteira espanhola na região onde o Douro
penetra em Portugal. No interior da zona de monotongação sobrevive, no
distrito de Leiria, uma pequena área em que o "ou" se conservou.>
Page 63:
<A monotongação de "ou" em /o/ é simétrica, a partir de uma data difícil de
determinar, à monotongação de "ei" em /e/*. Mas este fenómeno, ao contrário
do primeiro, não foi admitido na língua comum. A razão é fácil de
compreender quando se estuda o limite que separa a zona onde a monotongação
ocorreu (Sul) daquela em que o ditongo foi mantido (Norte). Esse limite é
uma linha que parte da costa atlântica, na altura de Óbidos, segue em
direcção ao sul passando por Lisboa, toma a direcção nordeste deixando ao
sul o vale do Tejo, acompanha o Zêzere até a fronteira espanhola, a nordeste
de Castelo Branco. Há, pois, mais uma vez, um Sul (zona inovadora) e um
Norte (zona conservadora). Mas a grande diferença em relação aos limites
dialectais precedentes é que, dessa feita, Lisboa está incluída na zona do
Norte.>
*Note the use of slashes. He is talking about phonemes. Consult any other
book on the subject and you'll find that the issue is anything but
controversial.
Regards,
Ekkehard
Ruud Harmsen
sci.lang:
>When peple spell their dialect with 'wiad' for 'wird',
I suppose nobody does. It is perceived as [r], because it is /r/, even
it usually isn't [r].
Ruud Harmsen
sci.lang:
>Ruud Harmsen wrote:
>>> Is there an [a] in german dialectal 'wiad' or 'woad'?
>>
>> (For <wird>?) My German-Dutch Van Dale dictionary write turned a
>> (Sampa [6]) for all final <r>'s.
>
>Is that the same you use for portuguese /@/?
(Assuming by /@/ you mean the vowel in written <a> vida, and cama).
I don't, but Sampa and Teach Yourself 1978 do.
> From a distance, I find german dialects so similar between themselves
>and to the standard, that I have trouble understanding why such a fuss
>is made about their differences. Isn't english as dialectally diverse,
>if we consider phonetics only?
In Britain, yes. In America, no.
António Marques
>> From a distance, I find german dialects so similar between themselves
>> and to the standard, that I have trouble understanding why such a fuss
>> is made about their differences. Isn't english as dialectally diverse,
>> if we consider phonetics only?
>
> In Britain, yes. In America, no.
I was allowing both plus SA, Canada and Oceania.
António Marques
>> Just because you can't hear it, it doesn't mean the difference doesn't
>> exist.
>
> Yes, that's exactly what it means, unless by "you" you mean me personally.
If I didn't mean you personally, it would be quite an enigmatic remark
now wouldn't it?
>>>> Is there an [a] in german dialectal 'wiad' or 'woad'?
>>> Sorry?
>> When peple spell their dialect with 'wiad' for 'wird', you might take
>> that as evidence that their wird has some kind of [a]. Just as you might
>> assume that if someone says they say 'lo^c,a' for 'louc,a', their word
>> is evidence that they have plain /o/ there; of course, if they then
>> proceed to say it's a longer o^ than the usual one, you can discount
>> that as spelling induced imagination and their word is no longer good.
>
> See below.
(I saw below, but saw nothing that addressed this particular point.)
>> It's really no big deal - you've decided that <ou> is /o/ and nothing
>> can make you change your mind. I don't have a problem with that, all I
>> care is to present the information I know so that others can be... more
>> informed.
>
> I appreciate you listen very closely to the way people speak, but I regret
> to say I'm getting the impression that you don't really know how to
> interpret what you hear.
It's not really a matter of exegesis. I know <ou> isn't plain /o/ much
the same as I know how to tell my d's from t's even when they're not at
all clear (and no, 'my' here doesn't mean they're produced by me).
> I can assure you I'm not confusing eye dialect with reality. If you don't
> want to take my word for it, let me suggest reading about the phonology of
> Portuguese and/or talking to Portuguese linguists about the subject. An
> excellent starting point would be Paul Teyssier's "História da Língua
> Portuguesa". Here's what he has to say about "ou" and "ei":
You have it? How good is it? I only read an abridged version.
> Page 52 (7th Portuguese edition, 1997):
> <O ditongo "ou", isto é, [ow], passou a [o] no actual português comum [...].
> Esta monotongação começou provavelmente a manifestar-se no século XVII.
> Invadiu todo o Sul e a maior parte do Centro de Portugal, mas no resto do
> país, ou seja, uma vez mais, no Norte, o antigo ditongo "ou" [ou] continua
> vivo. O limite do fenómeno parte do oeste da região de Aveiro e forma uma
> espécie de S que alcança a fronteira espanhola na região onde o Douro
> penetra em Portugal. No interior da zona de monotongação sobrevive, no
> distrito de Leiria, uma pequena área em que o "ou" se conservou.>
Which is copy+paste, if I'm not mistaken, of isogloss maps sloppily
drawn in the 1960s and never put to any quality checking. *However*,
that line may well describe where <ou> ceases to have a distinct [w]. As
I've said countless times, that isn't what I'm talking about.
> Page 63:
> <A monotongação de "ou" em /o/ é simétrica, a partir de uma data difícil de
> determinar, à monotongação de "ei" em /e/*. Mas este fenómeno, ao contrário
> do primeiro, não foi admitido na língua comum. A razão é fácil de
> compreender quando se estuda o limite que separa a zona onde a monotongação
> ocorreu (Sul) daquela em que o ditongo foi mantido (Norte). Esse limite é
> uma linha que parte da costa atlântica, na altura de Óbidos, segue em
> direcção ao sul passando por Lisboa, toma a direcção nordeste deixando ao
> sul o vale do Tejo, acompanha o Zêzere até a fronteira espanhola, a nordeste
> de Castelo Branco. Há, pois, mais uma vez, um Sul (zona inovadora) e um
> Norte (zona conservadora). Mas a grande diferença em relação aos limites
> dialectais precedentes é que, dessa feita, Lisboa está incluída na zona do
> Norte.>
>
> *Note the use of slashes. He is talking about phonemes. Consult any other
> book on the subject and you'll find that the issue is anything but
> controversial.
Note the epistemological convenience: not only it fits nicely with
preconceived ideas about phonological change (somewhere in there you'll
find that <ou> -> /o/ because when 'adorar' became the same as 'a
durar', its slot was made free for 'a dourar'), there's even a great
explanation for why <ei> -> /e/ is more restricted, that it didn't make
it to Lisbon. Never mind that Lisbon does exhibit a larger degree of ei
monphtongation than, say, Coimbra.
António Marques
>> Just because <ou> isn't [ow] south of, say, S. Pedro do Sul (which, for
>> the uninitiated, is to the NW of Viseu), that doesn't mean it merges
>> with <o^>. It's always a tad bit more rounded, or longer, or more
>> closed, or the 3. I admit it may be difficult to hear, but that's the
>> way it is, it's not may fault.
>
> I always had the feeling there was some difference too, without being
> able to say what. I don't know how much of that is due to expectation
> from the spelling.
From my experience, the spelling isn't always reliable, erring both
ways, though not much. But it's a muddy territory anyhow.
> I knew the spelling long before I read about (=
> didn't myself discover by listening) the phonetic similarity or
> identity of ou and ô in Portuguese.
Well, it's not like [ow] isn't somewhat similar to [o:] no matter the
language.
>> (*) Let alone that laminal sibilants affect the entire NE, not Viseu in
>> particular.
>
> Meaning there is a difference between paço and passo? Vianna mentioned
> that in 1892, for Camões' time and today's (his today, that is)
> Trás-os-Montes. http://purl.pt/146, pages 90-93.
And in fact that's more than a century ago. I've tried to investigate
what the situation might be nowadays, but haven't gotten conclusive data
yet.
c, z ss s
--------------------------------------
PI ts dz s. z.
MP,CPt s z s. z.
CP s z s z
CPb s. z. s. z.
CGg T D s. z.
CG, CS T T s. s.
(One place's [s.] isn't necessarily the same as other place's [s.])
PI Proto-Iberian
MP Medieval Portuguese
CPt 1892 Tras-os-Montes
CP Contemporary Portuguese
CPb Contemporary Viseu, NE Beira and probably N and E of that
CGg Contemporary Galician near the Gere^s
CG Contemporary Galician
CS Contemporary Spanish
Of course, other variants exist. These are the more interesting ones.
And relating to my remark, most people from Coimbra and south of that
will readily interpret [s.] as [S].
> ===
> 70. Parece averiguado que há três séculus a pronuncia [sic, não
> "pronúncia"; RH] do português de Lisbôa differia da actual nos
> sequintes pontos, que deveriam ser tidos em attenção numa leitura
> rigorosa do poêma.
Oh, the vagrancies of the good old orthography.
. /z/ was <z> a lot more times, usually whenever there wasn't a 'minimal
orthographical pair' with <s>
. OTOH learned words gleefully used <s>, such as in the -isar suffix.
. Accents were liberally used to distinguish quality in plain words, but
seldom used at all in learned ones
. Oddly enough, even with all the double consonants, some of today's
silent ones wre missing
> I. O s inicial e ss mediaes (saber, passo) differençavam-se do ç ou c
> antes de e, i, em que, como ainda hoje em Trás-os-Montes e parte do
> Minho e Beiras, eram preferidos com a superficie anterior do ápice da
> lingua, aproximando esse ápice, assim cóncavo, das gengivas dos
> incisivos superiores, posição que denominámos reversa, o que indicámos
> pelo sýmbolo s [com ponto debaixo] isto é, [s longo]. Êste valor do s
> mantinha-se-lhe depois de consoante, quando final na pausa, e antes
> das consoantes surdas p, t, c, qu, ç, f, x. Portanto o vocábulo p a s
> s o era differente de p a ç o, êstes pronunciava-se êstes [não consio
> lê-lo exactamente] (V.42).
[pas.u] vs [pasu], I suppose.
(Old [s.] vs [ts])
> II. Semelhantemente, z inicial ou medial differençava-se de s sonoro
> entre vogaes, em que êste era, como é em Trás-os-Montes e parte do
> Minho, um z proferido com os órgãos a mesma posição que fica descrita
> para o s inicial (V. p. 47), distinguindo-se conseguintemente o
> vocábulo c o s e r ([?????]) do vocábulo c o z e r.
([?????]) ?
[kuz.er] vs [kuzer]?
(Old [z.] vs [tz])
> O z final, porém, assim como a z interno, muito raro, antes de
> consoante surda, proferia-se, o que acontece actualmente em
> Trás-os-Montes, como ç; assim paz, luz pállida, luz azul, luz verde,
> pronunciavam-se respectivamente: paç, luçpálida, luzazúl, luz(z)vêrde.
Still some decades ago in parts of the Viseu district.
Ekkehard Dengler
António Marques schrieb:
> Ekkehard Dengler wrote:
> >> When peple spell their dialect with 'wiad' for 'wird', you might take
> >> that as evidence that their wird has some kind of [a]. Just as you
might
> >> assume that if someone says they say 'lo^c,a' for 'louc,a', their word
> >> is evidence that they have plain /o/ there; of course, if they then
> >> proceed to say it's a longer o^ than the usual one, you can discount
> >> that as spelling induced imagination and their word is no longer good.
> >
> > See below.
>
> (I saw below, but saw nothing that addressed this particular point.)
If an excerpt from "História da Língua Portuguesa" is not the kind of
evidence you're prepared to accept, then I rest my case.
> > I appreciate you listen very closely to the way people speak, but I
regret
> > to say I'm getting the impression that you don't really know how to
> > interpret what you hear.
>
> It's not really a matter of exegesis. I know <ou> isn't plain /o/ much
> the same as I know how to tell my d's from t's even when they're not at
> all clear (and no, 'my' here doesn't mean they're produced by me).
I don't mean to offend you, but I still suspect your judgment may be
influenced either by your own pronunciation or by the regional accent you're
most familiar with. Here's an idea: Have 100 southerners pronounce either
"este tipo de coro" or "este tipo de couro" for you and tape their
pronunciations. Jumble them if you can. When you play back the recording,
you won't be able to tell which is which. Nor will they. Alternatively,
switch on the television and listen to how most people speak.
> > I can assure you I'm not confusing eye dialect with reality. If you
don't
> > want to take my word for it, let me suggest reading about the phonology
of
> > Portuguese and/or talking to Portuguese linguists about the subject. An
> > excellent starting point would be Paul Teyssier's "História da Língua
> > Portuguesa". Here's what he has to say about "ou" and "ei":
>
> You have it? How good is it?
I have it, and I think it's good.
> > Page 52 (7th Portuguese edition, 1997):
> > <O ditongo "ou", isto é, [ow], passou a [o] no actual português comum
[...].
> > Esta monotongação começou provavelmente a manifestar-se no século XVII.
> > Invadiu todo o Sul e a maior parte do Centro de Portugal, mas no resto
do
> > país, ou seja, uma vez mais, no Norte, o antigo ditongo "ou" [ou]
continua
> > vivo. O limite do fenómeno parte do oeste da região de Aveiro e forma
uma
> > espécie de S que alcança a fronteira espanhola na região onde o Douro
> > penetra em Portugal. No interior da zona de monotongação sobrevive, no
> > distrito de Leiria, uma pequena área em que o "ou" se conservou.>
>
> Which is copy+paste, if I'm not mistaken, of isogloss maps sloppily
> drawn in the 1960s and never put to any quality checking. *However*,
> that line may well describe where <ou> ceases to have a distinct [w]. As
> I've said countless times, that isn't what I'm talking about.
The following sentence makes it quite clear that Teyssier is talking about
phonemes.
Yes, you'll find it in the second passage I quoted (p. 63). I repeat:
Consult *any* other book. I must say I'm finding your comments on this
subject increasingly perplexing. I'm reminded of native speakers of German
who claim that "Fälle" and "Felle" aren't homophonous. At any rate, there
doesn't seem to be much point in taking this discussion any further.
> Never mind that Lisbon does exhibit a larger degree of ei
> monphtongation than, say, Coimbra.
I can't really judge that, but I certainly don't think that [e] in words
like "beira" is a typical feature of the Lisbon accent. As I'm sure you
know, the current standard pronunciation of "ei", i.e. [@i], was originally
confined to Lisbon, where today the first element of the diphthong can
approach [a]. Or as Teyssier puts it: "Sabemos que foi em Lisboa que se deu
o primeiro aparecimento desse [@y]*, que, naturalmente, é desconhecido na
zona meridional de monotongação."
*I've changed the notation.
Regards,
Ekkehard
Ruud Harmsen
<ED...@t-online.de>: in sci.lang:
>Here's an idea: Have 100 southerners pronounce either
>"este tipo de coro" or "este tipo de couro" for you and tape their
>pronunciations. Jumble them if you can. When you play back the recording,
>you won't be able to tell which is which. Nor will they.
A few weeks ago I did a similar test with sips of a reasonable real
Port wine, and of a cheap French imitation. Result: 7 times right, 5
times wrong. Not statistically relevant, conclusion: I can't taste the
difference. Yet, without the blindfold, I was and am convinced that
there is a marked difference in smell and taste.
What one expects to taste/see/hear has a huge influence on what one
sincerely tastes/sees/hears etc. That shouldn't be underestimated.
António Marques
>> Here's an idea: Have 100 southerners pronounce either "este tipo de
>> coro" or "este tipo de couro" for you and tape their
>> pronunciations. Jumble them if you can. When you play back the
>> recording, you won't be able to tell which is which. Nor will they.
>
> A few weeks ago I did a similar test with sips of a reasonable real
> Port wine, and of a cheap French imitation. Result: 7 times right, 5
> times wrong. Not statistically relevant,
Hey, you can't do that. You have to guess the distribution family, its
parameters, do some horribly uninteresting math and then you'll know how
much times are statistically relevant. And then it may be that the cheap
imitation was a particularly good cheap imitation, and the reasonable
port wasn't particularly good reasonable port. And 'port' doesn't have
capital P.
As for the orginal experiment, you'd have to correct it: record the
*same* person say both things in normal discourse over the day. Then
pick only the minimal pairs and see if you can tell the difference.
That'll inform you about that speaker and yourself (other people may get
worse or better results).
> conclusion: I can't taste the difference. Yet, without the blindfold,
> I was and am convinced that there is a marked difference in smell and
> taste.
>
> What one expects to taste/see/hear has a huge influence on what one
> sincerely tastes/sees/hears etc. That shouldn't be underestimated.
It's like that experiment where they colour the same candy with
different dyes, and then people say the orange ones taste like orange,
the yellow ones like lemon and the red ones like strawberry. Or else
that's an urban myth.
António Marques
>>>> assume that if someone says they say 'lo^c,a' for 'louc,a',
>>>> their word is evidence that they have plain /o/ there; of
>>>> course, if they then proceed to say it's a longer o^ than the
>>>> usual one, you can discount that as spelling induced
>>>> imagination and their word is no longer good.
>>>
>>> See below.
>>
>> (I saw below, but saw nothing that addressed this particular
>> point.)
>
> If an excerpt from "História da Língua Portuguesa" is not the kind of
> evidence you're prepared to accept, then I rest my case.
If it were it would have worked when I read the book, no?
(But this comment of yours belongs below, not here. Here I was pointing
out how some kinds of evidence can be gleefully or uncounsciously
manipluated.)
>> It's not really a matter of exegesis. I know <ou> isn't plain /o/
>> much the same as I know how to tell my d's from t's even when
>> they're not at all clear (and no, 'my' here doesn't mean they're
>> produced by me).
>
> I don't mean to offend you, but I still suspect your judgment may be
> influenced either by your own pronunciation or by the regional
> accent you're most familiar with. Here's an idea: Have 100
> southerners pronounce either "este tipo de coro" or "este tipo de
> couro" for you and tape their pronunciations. Jumble them if you can.
> When you play back the recording, you won't be able to tell which is
> which. Nor will they. Alternatively, switch on the television and
> listen to how most people speak.
You seem firm on the assumption that I'm saying that <ou> is [ow].
(Apart from the north,) it's not. Nor is it in my own accent(^endnote).
>> Which is copy+paste, if I'm not mistaken, of isogloss maps sloppily
>> drawn in the 1960s and never put to any quality checking.
>> *However*, that line may well describe where <ou> ceases to have a
>> distinct [w]. As I've said countless times, that isn't what I'm
>> talking about.
>
> The following sentence makes it quite clear that Teyssier is talking
> about phonemes.
Indeed when dicussing <ei>; and by that, implicitely, in the discussion
of <ou>. But notice no mention is made of plain <o> or <e>, and one was
called for even if it were to say the diphtongs have coalesced with the
plain versions, again even if one's supposed to assume that that's
precisley what he intended, and meant by /e/ and /o/. That's a bit of
sloppy style. I thought it was due to the abridging, but it seems it's
on the full version as well.
However, the matter remains that if the S line is supposed to delimit
full [ow] (as I think it does), the rest of the country isn't homogenous
and either the 'monophtongation area' includes areas where monophtongal
<ou> is different from <o> and he should have expanded, or he's
misinformed; if the S line is supposed to delimit where <ou> regardless
of pronunciation is separate from <o>, then it's just wrong, because
<ou> is mostly [ow] inside it, and is separate from <o> south of the
line even if it weren't all the way to Lisbon.
>> Note the epistemological convenience: not only it fits nicely with
>> preconceived ideas about phonological change (somewhere in there
>> you'll find that <ou> -> /o/ because when 'adorar' became the same
>> as 'a durar', its slot was made free for 'a dourar'), there's even
>> a great explanation for why <ei> -> /e/ is more restricted, that it
>> didn't make it to Lisbon.
>
> Yes, you'll find it in the second passage I quoted (p. 63).
I know, I was chiding it.
> I repeat: Consult *any* other book. I must say I'm finding your
> comments on this subject increasingly perplexing. I'm reminded of
> native speakers of German who claim that "Fälle" and "Felle" aren't
> homophonous. At any rate, there doesn't seem to be much point in
> taking this discussion any further.
Or dass and das have different vowels, right? But I'm not saying that
c-cedille is different from ss, or that Azerbaijao is different from
Azerbeijao, or papeis from pape´is, or mem from ma~e, or vejo from beijo
for any but some speakers/regions.
>> Never mind that Lisbon does exhibit a larger degree of ei
>> monphtongation than, say, Coimbra.
>
> I can't really judge that, but I certainly don't think that [e] in
> words like "beira" is a typical feature of the Lisbon accent. As I'm
> sure you know, the current standard pronunciation of "ei", i.e. [@i],
> was originally confined to Lisbon,
The Lisbon area exhibited some degree of [@j] -> [@] (afecting both <ei>
and <ej>, which has a glide due to the /Z/), which hasn't decided
whether it's standard, substandard or posh. Meanwhile, [@j] hasn't been
droven out of its position.
> where today the first element of the diphthong can approach [a]. Or
> as Teyssier puts it: "Sabemos que foi em Lisboa que se deu o primeiro
> aparecimento desse [@y]*, que, naturalmente, é desconhecido na zona
> meridional de monotongação."
Inside which, indeed, one may still find full [ej] in a couple of
situations, sometimes. [O r:@:m@ d.@ oliv.ejr@]... (where . marks some
unspecified phonetic phenomenon).
(^) The most distinctive features of Coimbra vs Lisbon are:
Lisbon Coimbra
--------------------------
elh @L eL
ej >@Z< @jZ
oj >oZ< ojZ
-io iw iu*
'><' indicate that it may not be standard but is found [mostly] there.
Of course, there are variants.
(*) the /i/ and /u/ do not form a diphtong, for whatever reason. I still
don't think that warrants a /w/.
As for me, for some reason:
- I have [O] in <o~e(s)>, which is decidedly substandard. Can't help it.
- /s/ is between [s] and [s.], though much closer to [s]. I have no
trouble switching to [s.], because the distance is greater; conversely,
I can't maintain a conversation with [s] because the distance is smaller
and harder to keep track of. However, it's enough that [s] people may
judge me as using [s.]. Otoh, when speaking english I use [s]. Go figure.
Ruud Harmsen
sci.lang:
>Hey, you can't do that. You have to guess the distribution family, its
>parameters, do some horribly uninteresting math and then you'll know how
>much times are statistically relevant. And then it may be that the cheap
>imitation was a particularly good cheap imitation, and the reasonable
>port wasn't particularly good reasonable port. And 'port' doesn't have
>capital P.
You're probably right, on all counts.
What remains interesting though, is that without the blindfold, the
taste and smell difference is clear. With it, it often isn't, and when
it is, it's not reliable.
Ruud Harmsen
>>> Here's an idea: Have 100 southerners pronounce either "este tipo de
>>> coro" or "este tipo de couro" for you and tape their
>>> pronunciations. Jumble them if you can. When you play back the
>>> recording, you won't be able to tell which is which. Nor will they.
Mon, 21 Nov 2005 00:18:44 +0000: António Marques <m....@sapo.pt>: in
sci.lang:
>As for the orginal experiment, you'd have to correct it: record the
>*same* person say both things in normal discourse over the day.
Also interestin, but not quite necessarily. When a phonetic difference
is discernable (or even phonetically relevant) in a given variety of a
language, it normally is that regardless of individual differences
between people.
>Then
>pick only the minimal pairs and see if you can tell the difference.
Yes, minimal pairs are required, otherwise the listeners won't be
impartial, and can hear things that aren't there.
>It's like that experiment where they colour the same candy with
>different dyes, and then people say the orange ones taste like orange,
>the yellow ones like lemon and the red ones like strawberry. Or else
>that's an urban myth.
It was a BBC article mentioned in a Dutch newsgroup, that led to my
experiment: http://news.bbc.co.uk/1/hi/health/4558075.stm
Ekkehard Dengler
António Marques schrieb:
> Ekkehard Dengler wrote:
>
> > influenced either by your own pronunciation or by the regional
> > accent you're most familiar with. Here's an idea: Have 100
> > southerners pronounce either "este tipo de coro" or "este tipo de
> > couro" for you and tape their pronunciations. Jumble them if you can.
> > When you play back the recording, you won't be able to tell which is
> > which. Nor will they. Alternatively, switch on the television and
> > listen to how most people speak.
>
> You seem firm on the assumption that I'm saying that <ou> is [ow].
I don't see how that would follow from what I wrote. No, if I understand you
correctly, you think that even monophthongal realisations of the phoneme
written <ou> are consistently differentiated from realisations of the /o/
phoneme. And I think you're completely wrong, that's all.
Regards,
Ekkehard
Peter T. Daniels
What would be most amusing would be to give the blindfolded wine-taster
the _very same glass_ a second time. (Perhaps immediately, perhaps with
some intervening.)
--
Peter T. Daniels gram...@att.net
Ruud Harmsen
<gram...@worldnet.att.net>: in sci.lang:
>What would be most amusing would be to give the blindfolded wine-taster
>the _very same glass_ a second time. (Perhaps immediately, perhaps with
>some intervening.)
We did, we used two pre-filled and pre-labelled glasses. They were
filled while I already had the blindfold on, so I couldn't identify
the smell. They were handed over to me at random, for a small sip, 12
times.
On time I complained that one of the glasses was less full than the
other one, so I could identify them that way. She admitted that, and
"corrected" the problem. But in hindsight, it wasn't true.
António Marques
>> You seem firm on the assumption that I'm saying that <ou> is [ow].
>
> I don't see how that would follow from what I wrote.
It followed from the beginning of the paragraph,
> I still suspect your judgment may be influenced either by your own
> pronunciation or by the regional accent you're most familiar with.
> Here's an idea: Have 100 southerners pronounce either (...)
My pronunciation and the ones I'm most familiar with do not have [ow],
but neither do they coalesce <ou> and <o>. As for the southerners, most
people you see on TV are from Lisbon, and I routinely go to Setubal for
a few days, so it's not like I don't have any familiarity with 'southerner'.
> No, if I understand you correctly, you think that even monophthongal
> realisations of the phoneme written <ou> are consistently
> differentiated from realisations of the /o/ phoneme.
(I don't think there is a phoneme written <ou>, I think the /o/ phoneme
followed by /u/ have a different realisation from plain /o/. You think
you can squeeze 'comigo' into 1 syllable by having the /u/s disappear
into a slight rounding of the preceding consonants, but seem unwilling
dich to vorstellen that a /u/ after an /o/ might disappear leaving only
a slightly rounder/longer/closer o.)
Ekkehard Dengler
António Marques schrieb:
> (I don't think there is a phoneme written <ou>, I think the /o/ phoneme
> followed by /u/ have a different realisation from plain /o/.
You don't mean to tell me that <ou> is actually disyllabic?
> You think
> you can squeeze 'comigo' into 1 syllable by having the /u/s disappear
> into a slight rounding of the preceding consonants, but seem unwilling
> dich to vorstellen that a /u/ after an /o/ might disappear leaving only
> a slightly rounder/longer/closer o.)
Imagination doesn't come into it at all. I'm talking about what I hear,
which very probably coincides with what every single linguist has ever said
about this.
Regards,
Ekkehard
António Marques
>> (I don't think there is a phoneme written <ou>, I think the /o/
>> phoneme followed by /u/ have a different realisation from plain
>> /o/.
>
> You don't mean to tell me that <ou> is actually disyllabic?
What makes you think I do? I just mentioned elsewhere a falling diphtong
made up of /i/ followed by /u/.
('vo^o' is usually dyssilabic.)
>> You think you can squeeze 'comigo' into 1 syllable by having the
>> /u/s disappear into a slight rounding of the preceding consonants,
>> but seem unwilling dich to vorstellen that a /u/ after an /o/ might
>> disappear leaving only a slightly rounder/longer/closer o.)
>
> Imagination doesn't come into it at all.
I think you are wrong.
> I'm talking about what I hear, which very probably coincides with
> what every single linguist has ever said about this.
And which (what you hear), applying the same rules, would probably be
different if you had come across other analyses. It's not like *many*
people have delved into it.
António Marques
>> When peple spell their dialect with 'wiad' for 'wird',
>
> I suppose nobody does. It is perceived as [r], because it is /r/, even
> it usually isn't [r].
(I suppose you've come across stuff such as
http://www.landluft.net/texte.htm)
Ruud Harmsen
sci.lang:
>But unstressed /ou/ is longer than stressed /o/, and even if it were as
>long, it would be the only portuguese phoneme that wouldn't be shorter
>when unstressed, which isn't a very useful analysis, and what do you do
>with de facto unstressed short [o], as in co-piloto, cofragem, etc? What
>phoneme is it? Those words don't sound like *cou-piloto or *coufragem.
Sometimes however, an original ou, when unstressed, develops into [u]
just like unstressed /o/ did:
Modern 'aposento' and 'aposentar' used to be 'apousento' and
apousentar'. Source of the latter: Porto Editora, source of the
latter: Luiz Camões, Canto I, LX. A modern orthographic edition (from
a booklet I bought in a Coimbra supermarket in 1999, for somethig like
390 escudos):
===
Partia, alegremente navegando,
A ver as naus ligeiras lusitanas,
Com refresco da terra, em si cuidando
Que são aquelas gentes inumanas
Que, os aposentos cáspios habitando,
A conquistar as terras asianas
Vieram, e, por ordem do Destino
O Império tomaram a Constantino.
===
But the 1572 original
http://purl.pt/1
http://purl.pt/1/2/
http://purl.pt/1/2/cam-3-p_PDF/cam-3-p_PDF_24-C-R0072/cam-3-p_0002_5r-11r_t24-C-R0072.pdf
has this:
===
Partia alegremente nauegando,
A ver as naos ligeiras Lusitanas,
Com refresco da terra, em si cuidando,
Que sam aquellas gentes inhumanas:
Que os apousentos Caspios habitando,
A conquistar as terras Asianas
Vierão: & por ordem do destino
O Imperio tomarão a Constantino.
===
Ruud Harmsen
<realemail...@rudhar.com.invalid>: in sci.lang:
>But the 1572 original
>http://purl.pt/1
>http://purl.pt/1/2/
>http://purl.pt/1/2/cam-3-p_PDF/cam-3-p_PDF_24-C-R0072/cam-3-p_0002_5r-11r_t24-C-R0072.pdf
In the original of this famous Portuguese , we sometimes see "German"
ß's, like in the 1st first verse, last line: De Assirios, written De
Aßirios.
2nd verse, last line but one: Aßi for what in modern editions is
'assi', and 'assim' in contemporary Portuguese. Cf. Spanish así.
In the same canto, however, on the left of page 11, the same word is
written 'assy', with two long s's and a Greek y.
Ruud Harmsen
<realemail...@rudhar.com.invalid>: in sci.lang:
>===
>Partia alegremente nauegando,
>A ver as naos ligeiras Lusitanas,
>Com refresco da terra, em si cuidando,
>Que sam aquellas gentes inhumanas:
>Que os apousentos Caspios habitando,
>A conquistar as terras Asianas
>Vierão: & por ordem do destino
>O Imperio tomarão a Constantino.
>===
Could we perhaps conclude from this, that the modern "contractions" of
European Portuguese (reduction of unstressed <o> and <e>) already
existed in 1572? If we read 'refresco' in the 3rd line above in a
somewhat Castillian/Galician/Brazilian Portuguese way, with full
vowels in the first and maybe last syllable, the rhythm/metrum or
whatever it is (I know very little about poetry) somehow doesn't work
out. But if read it in the modern EP way, in one syllable, rfresc
[R\freSk<sup>w</sup>], it fits much better, doesn't it?
Also the last a of partia and the first of alegremente need to be one
vowel (this is also the case in Modern Spanish) to make it work.
andrew...@hotmail.com
I know nothing about specifically Portuguese versification (though I
have read the Lusiads), but surely line 3 is a standard
hendecasyllable if you have synalepha of
(ter)ra_em ? No need for other vowel reduction.