Fri, 4 Jun 2021 10:58:40 -0700 (PDT): "Peter T. Daniels"
<grammatim@verizon.net> scribeva:

>On Friday, June 4, 2021 at 11:54:33 AM UTC-4, Ruud Harmsen wrote:
>> Fri, 4 Jun 2021 01:07:28 -0700 (PDT): Dingbat
>> <ranjit_...@yahoo.com> scribeva:
>
>> >I find this word enigmatic:
>> >encastré
>> >https://forvo.com/word/encastr%C3%A9/
>>
>> >The terminal vowel sounds like [I]!
>> > (the ending vowel of ancestry)
>
>(In early-20th-century RP. I don't know where Indian English stands
>on that suffix.)
>
>> Not surprising. The phonetic difference between [e] and [I] is largely
>> undefined.
>
>What an odd thing to say. If those two phones belong to different
>phonemes in one's native language, one has no trouble distinguishing
>them.

I asked about the phonetic difference.

>(Anyway you probably meant [E] and [I]?)

No, I did not.

>For [e]/[I], mate/mitt in English; for [E]/[I], met/mitt.

Yes, thanks.

>> I clearly hear a cardinal [e], however, as expected.
>
>You need to ask Ranjit to lay out the Malayalam vowel system
>to see why he hears what he hears.

Yes, probably. Or read up in Wikipedia:
https://en.wikipedia.org/wiki/Malayalam#Vowels
It has only close, mid and open. So [e], [E] and [I] is probably all
one phoneme.
--
Ruud Harmsen, http://rudhar.com

Fri, 4 Jun 2021 21:15:56 +0200: mabel wugi <wugi@scarlet.be> scribeva:

>Op 4/06/2021 om 19:58 schreef Peter T. Daniels:
>
>>> Not surprising. The phonetic difference between [e] and [I] is largely
>>> undefined.
>
>> What an odd thing to say. If those two phones belong to different
>
>I agree with Ruud.
>
>> phonemes in one's native language, one has no trouble distinguishing
>> them. (Anyway you probably meant [E] and [I]?)
>
>No.
>
>> For [e]/[I], mate/mitt in English; for [E]/[I], met/mitt. long in meer.
>
>
>For [e]/[I], smeer/smeed/smid in Dutch. [I] (very) short in smid, [e.]
>half long in smeed, [e:] long in smeer. Same mouth position each vowel.

For smeer and smid, yes. Smeed however seems to have a different
position, even if I reduce/eliminate my tendency towards a diphthong
there. But I don't know what it is.
--
Ruud Harmsen, http://rudhar.com

Fri, 4 Jun 2021 12:35:51 -0700 (PDT): "Peter T. Daniels"
<grammatim@verizon.net> scribeva:

>On Friday, June 4, 2021 at 3:16:02 PM UTC-4, wugi wrote:
>> Op 4/06/2021 om 19:58 schreef Peter T. Daniels:
>
>DO NOT DELETE ATTRIBUTIONS
>
>> >> Not surprising. The phonetic difference between [e] and [I] is largely
>> >> undefined.
>> > What an odd thing to say. If those two phones belong to different
>>
>> I agree with Ruud.
>
>Maybe he is misusing "undefined." What do you think he meant by it?

A clear discription of what the difference is, in acoustic,
articulary, auditory terms. Height, positition on front-back scale,
formants.

>> > phonemes in one's native language, one has no trouble distinguishing
>> > them. (Anyway you probably meant [E] and [I]?)
>>
>> No.
>
>[e] and [I] are quite different
>
>> > For [e]/[I], mate/mitt in English; for [E]/[I], met/mitt. long in meer.
>>
>> For [e]/[I], smeer/smeed/smid in Dutch. [I] (very) short in smid, [e.]
>> half long in smeed, [e:] long in smeer. Same mouth position each vowel.
>
>If those are phonemically different, there is no reason for Ruud
>to find them hard to distinguish.

Smeer and smeed aren't phonemically different, only phonetically (for
me, probably not for Wugi), conditioned by syllable-final /r/.
--
Ruud Harmsen, http://rudhar.com

On Friday, June 4, 2021 at 5:30:07 PM UTC-4, Christian Weisgerber wrote:
> On 2021-06-04, Peter T. Daniels <gram...@verizon.net> wrote:

> > [e] and [I] are quite different
>
> You need to listen to some New Zealanders. Their DRESS vowel is
> described as [e] or raised [e], but is frequently mistaken for the
> KIT vowel by AmE speakers.

That's because the phones are assigned to different phonemes in
the two dialects, not because of the amount of inherent similarity
or difference of the phones.

It normally takes a few seconds, if that long, for speakers of different
dialects to latch onto the somewhat different assignment of phonemes.

One of Labov's favorite tricks is to chop a sound (usually a vowel!) or
even a whole word out an utterance and ask a listener to identify it.
Usually either they can't or they guess wrong, but upon hearing one or
two seconds before and after the mystery sound, it's perfectly clear.

I've only just had to look at African phonologival phenomena. Turns
out that the feature "+ATR" was discovered only in the 1960s -- because
the vowels that in Bantu and many West African languages were transcribed
with [i I e E u U o O] -- in English, the "tense-lax" or "higher-lower" pairs, do
not in fact have any change in muscle tension or tongue height at all, but
instead degree of constriction at the root of the tongue. (The "A" is for Advanced.)

> Obligatory sketch from _Flight of the Concords_:
> https://www.youtube.com/watch?v=NRdg1MOYxHo

On Saturday, June 5, 2021 at 7:54:35 AM UTC-4, Dingbat wrote:
> On Friday, June 4, 2021 at 10:58:42 AM UTC-7, Peter T. Daniels wrote:
> > On Friday, June 4, 2021 at 11:54:33 AM UTC-4, Ruud Harmsen wrote:
> > > Fri, 4 Jun 2021 01:07:28 -0700 (PDT): Dingbat
> > > <ranjit_...@yahoo.com> scribeva:
> > > >I find this word enigmatic:
> > > >encastré
> > > >https://forvo.com/word/encastr%C3%A9/
> > > >The terminal vowel sounds like [I]!
> > > > (the ending vowel of ancestry)
> > (In early-20th-century RP. I don't know where Indian English stands
> > on that suffix.)
> > > Not surprising. The phonetic difference between [e] and [I] is largely
> > > undefined.
> > What an odd thing to say. If those two phones belong to different
> > phonemes in one's native language, one has no trouble distinguishing
> > them. (Anyway you probably meant [E] and [I]?)
>
> He couldn't have meant [E]. It is too far from [I] to be identified as [I].

Sorry, but in the American English dialect of St. Louis, Missouri (and
environs), "pin" and "pen" have merged. [I] and [E] are similar enough
to have become indistinguishable before [n]. (People there write with
an inkpen. Chicagoans find that amusing.)

> > For [e]/[I], mate/mitt in English; for [E]/[I], met/mitt.
> > > I clearly hear a cardinal [e], however, as expected.
>
> mate has [eI], not [e].

Depends on your phonemic analysis. British phoneticians prefer
to write [e:].

> > You need to ask Ranjit to lay out the Malayalam vowel system
> > to see why he hears what he hears.
>
> The Malayalam short e is slightly more open than cardinal e on a chart I've seen.
> The German [I] is slightly more closed than cardinal e on a chart I've seen.
> Both are retracted from the fronted position of cardinal e.
> The two are equidistant from cardinal e as far as I can tell.
> I don't remember an English chart but English doesn't have an [e].
> I've heard it from Anglophones only in a French word like
> DEPARTMENT, the political division in France, and not all
> Anglophones manage to make it a pure vowel.

Fri, 4 Jun 2021 21:02:02 -0000 (UTC): Christian Weisgerber
<naddy@mips.inka.de> scribeva:

>On 2021-06-04, Peter T. Daniels <grammatim@verizon.net> wrote:
>
>> [e] and [I] are quite different
>
>You need to listen to some New Zealanders. Their DRESS vowel is
>described as [e] or raised [e], but is frequently mistaken for the
>KIT vowel by AmE speakers.

Yes. Their sax is sex, their sex is six, and their six sucks.
Some kind of musical orgy that wasn't a good idea after all.

>Obligatory sketch from _Flight of the Concords_:
>https://www.youtube.com/watch?v=NRdg1MOYxHo

--
Ruud Harmsen, http://rudhar.com

On Saturday, June 5, 2021 at 8:33:06 AM UTC-4, Arnaud Fournet wrote:
> Le vendredi 4 juin 2021 à 14:21:15 UTC+2, wugi a écrit :
> > Op 4/06/2021 om 11:55 schreef António Marques:

> > > In the french I was taught, <ai> was just the same as <è>. I've since been
> > > made aware that is often the same as <é>, but not the rules for that. This
> > > looks like as good as time as any other to inquire about those.
> > > What are the rules? Quelles sont-elles, les règles?
> > There aren't clear ones, and if there are, they're not universal. From
> > what there is, the distinction is better maintained in eg. Belgium
> > (except dialectical) than in 'mainstream' Île de France (which is also
> > in the stage of coinciding -in and -un).
> > -ai ~ é: parlai ~ parlé, parler, parlez etc.
> > -ais, -ait, -aient ~ è: mais ~ mets
> > -aise ~ è: (ma)laise ~ (mé)lèze.
>
> It's the theoretical phonology of Parisian French,
> but I seriously doubt present-day speakers abide by these laws.

In school French class in the early 1960s, they (our teacher was
a native speaker) told us about that distinction (and also pâte/patte)
but said you'd only hear it at the Comédie Française.

That was more than 50 years ago.

> > But the increased confusion é,è shows up in spelling errors between
> > parlai and parlais, parlerai and parlerais (besides between homophonic
> > parlé, parler and parlez). Also in spelling confusion éléverai/élèverai;
> > accéderai/accèderai....
>
> There's a lot of confusion in actual speech and idiolects.

Plus ça change, plus c'est la même cause.

On Saturday, June 5, 2021 at 9:40:04 AM UTC-4, Ruud Harmsen wrote:
> Fri, 4 Jun 2021 12:35:51 -0700 (PDT): "Peter T. Daniels"
> <gram...@verizon.net> scribeva:
> >On Friday, June 4, 2021 at 3:16:02 PM UTC-4, wugi wrote:
> >> Op 4/06/2021 om 19:58 schreef Peter T. Daniels:
> >
> >DO NOT DELETE ATTRIBUTIONS
> >
> >> >> Not surprising. The phonetic difference between [e] and [I] is largely
> >> >> undefined.
> >> > What an odd thing to say. If those two phones belong to different
> >> I agree with Ruud.
> >Maybe he is misusing "undefined." What do you think he meant by it?
>
> A clear discription of what the difference is, in acoustic,
> articulary, auditory terms. Height, positition on front-back scale,
> formants.

Perhaps in your autodidacticism you've never encountered the prolific
authors Peter Ladefoged and Ian Maddieson. They provide charts
galore of the formant-based vowel triangles of English and many other
languages averaged from dozens (or more; fewer for endangered languages,
of course) of speakers.

> >> > phonemes in one's native language, one has no trouble distinguishing
> >> > them. (Anyway you probably meant [E] and [I]?)
> >> No.
> >[e] and [I] are quite different
> >> > For [e]/[I], mate/mitt in English; for [E]/[I], met/mitt. long in meer.
> >> For [e]/[I], smeer/smeed/smid in Dutch. [I] (very) short in smid, [e.]
> >> half long in smeed, [e:] long in smeer. Same mouth position each vowel.
> >If those are phonemically different, there is no reason for Ruud
> >to find them hard to distinguish.
>
> Smeer and smeed aren't phonemically different, only phonetically (for
> me, probably not for Wugi), conditioned by syllable-final /r/.

Oy. Sounds like the British r-less problem -- they have to posit a whole
extra series of surface phonemes to account for the vowels before
former /r/. (See both ita and Shavian, which you can compare side by
side in my *Exploration*.)

On 2021-06-05, Dingbat <ranjit_mathews@yahoo.com> wrote:

> I don't remember an English chart but English doesn't have an [e].

At that level of detail you need to specify which variant of English.
Southern hemisphere accents have raised the DRESS vowel to [e].

--
Christian "naddy" Weisgerber naddy@mips.inka.de

On 2021-06-04, António Marques <antonioprm@sapo.pt> wrote:

> In the french I was taught, <ai> was just the same as <è>. I've since been
> made aware that is often the same as <é>, but not the rules for that. This
> looks like as good as time as any other to inquire about those.
> What are the rules? Quelles sont-elles, les règles?

That is part of the bigger question how the mid-close /e, ø, o/ and
mid-open vowels /ɛ, œ, ɔ/ are distributed. The answer to that is
made complicated by regional differences--don't trust the Belgian!--and
the fact that the descriptive rules from 1900 have fossilized into
prescriptive ones.

Based on Fagyal/Kibbee/Jenkins, _French: A Linguistic Introduction_,
I'll try to summarize for Northern Metropolitan (= Parisian) French.

The basic rule is that CVC syllables have /ɛ, œ, ɔ/ and CV syllables
have /e, ø, o/. There are significant exceptions in word-final
syllables:

There is no /e/ in CVC syllables, but in CV syllables, an opposition
between /e/ and /ɛ/ is somewhat maintained. This is largely indicated
by the spelling where historically closed syllables like -et, -ait,
-ais have /ɛ/ and historically open ones like -é, -ai have /e/.
This distinction is breaking down and "appears to be reliably
maintained only in [...] frequent lexical items"; in particular the
conditional and future tense endings -rais and -rai have effectively
merged.

For /œ/-/ø/ and /ɔ/-/o/ the situation is reversed: There is no /œ/
or /ɔ/ in CV syllables. There is one regular modification for CVC
syllables: final /z/ forces /o/ and /ø/. Apart from that, the
opposition is somewhat maintained in CVC syllables. For /ø/ vs.
/œ/ the examples are very strained: veule - veulent, jeûne - jeune.
There are more potential minimal pairs for /o/ vs. /ɔ/, where /o/
is frequently spelled -ô- or -au-.

For non-final syllables the mid-open/mid-close distinction is
neutralized. Here vowel harmony can be frequently observed where
the mid vowels assimilate to the height of the final vowel.

I think overall it is fairly clear that the system is in a state
of collapse and that the formerly phonemic distinctions [e]-[ɛ],
[ø]-[œ], [o]-[ɔ] are well into the process of becoming allophonic.

----

Two further vowel mergers are worth mentioning:

The dictionary opposition /a/-/ɑ/ is dead. As recent editions of
Grevisse point out exasperatedly, even among the speakers who
maintain the distinction, there is too little agreement on which
word has which vowel.

The two nasal vowels /ɛ̃/ and /œ̃/ have merged in Parisian French.
Notice for instance the widespread cutesy abbreviation "1fo" for
"info". This seems to have crept up on everybody largely unnoticed,
since I think far less ink has been spilled over it than over the
other vowel changes.

--
Christian "naddy" Weisgerber naddy@mips.inka.de

On Saturday, June 5, 2021 at 6:34:42 AM UTC-7, Ruud Harmsen wrote:
> Fri, 4 Jun 2021 10:58:40 -0700 (PDT): "Peter T. Daniels"
> <gram...@verizon.net> scribeva:
> >On Friday, June 4, 2021 at 11:54:33 AM UTC-4, Ruud Harmsen wrote:
> >> Fri, 4 Jun 2021 01:07:28 -0700 (PDT): Dingbat
> >> <ranjit_...@yahoo.com> scribeva:
> >
> >> >I find this word enigmatic:
> >> >encastré
> >> >https://forvo.com/word/encastr%C3%A9/
> >>
> >> >The terminal vowel sounds like [I]!
> >> > (the ending vowel of ancestry)
> >
> >(In early-20th-century RP. I don't know where Indian English stands
> >on that suffix.)
> >
> >> Not surprising. The phonetic difference between [e] and [I] is largely
> >> undefined.
> >
> >What an odd thing to say. If those two phones belong to different
> >phonemes in one's native language, one has no trouble distinguishing
> >them.
> I asked about the phonetic difference.
> >(Anyway you probably meant [E] and [I]?)
> No, I did not.
> >For [e]/[I], mate/mitt in English; for [E]/[I], met/mitt.
> Yes, thanks.
> >> I clearly hear a cardinal [e], however, as expected.
> >
> >You need to ask Ranjit to lay out the Malayalam vowel system
> >to see why he hears what he hears.
> Yes, probably. Or read up in Wikipedia:
> https://en.wikipedia.org/wiki/Malayalam#Vowels
> It has only close, mid and open. So [e], [E] and [I] is probably all
> one phoneme.

Phonetically, there are 7 front vowels:
[i/I], [i:], [e],[e:], [E], [E:] and [&:].
The last is found in loans from English but there's no proper way to spell it.

The phonemes of which these are allophones::
Phoneme - Allophones
/i/ - [i],[I],[e] The last is in a /iCa/ context, eg. /ila/ realized as [ela].
/e/ - [e]
/a/ - [@],[a],[E] The last is like Arabic's imala realization of /a/.
/i:/ - [i:]
/e:/ - [e:],[E:] The latter is in an /e:r/ context, eg. /ve:re:/ realized as [wE:r.e:]
/&/ - [&:] long like in <jam> even in a word like <bat> where Brits have short [&].

P.S. My [a] means what Kirshenbaum calls [V].

On Saturday, June 5, 2021 at 7:30:06 AM UTC-7, Christian Weisgerber wrote:
> On 2021-06-05, Dingbat <ranjit_...@yahoo.com> wrote:
>
> > I don't remember an English chart but English doesn't have an [e].
> At that level of detail you need to specify which variant of English.
> Southern hemisphere accents have raised the DRESS vowel to [e].

DRESS has [e] in Malayalis' English also but I wasn't counting such accents.
The only thing I remember from an Oz English chart is that /aI/ is realized as [EI].

On Saturday, June 5, 2021 at 8:30:06 AM UTC-7, Christian Weisgerber wrote:
> On 2021-06-04, António Marques <anton...@sapo.pt> wrote:
>
> > In the french I was taught, <ai> was just the same as <è>. I've since been
> > made aware that is often the same as <é>, but not the rules for that. This
> > looks like as good as time as any other to inquire about those.
> > What are the rules? Quelles sont-elles, les règles?
> That is part of the bigger question how the mid-close /e, ø, o/ and
> mid-open vowels /ɛ, œ, ɔ/ are distributed. The answer to that is
> made complicated by regional differences--don't trust the Belgian!--and
> the fact that the descriptive rules from 1900 have fossilized into
> prescriptive ones.
>
> Based on Fagyal/Kibbee/Jenkins, _French: A Linguistic Introduction_,
> I'll try to summarize for Northern Metropolitan (= Parisian) French.
>
> The basic rule is that CVC syllables have /ɛ, œ, ɔ/ and CV syllables
> have /e, ø, o/. There are significant exceptions in word-final
> syllables:
>
> There is no /e/ in CVC syllables, but in CV syllables,

I've heard an Anglophone pronounce DEPARTMENT as
[depAtma]. Would the French find the [e] wrong?

> an opposition
> between /e/ and /ɛ/ is somewhat maintained.
> This is largely indicated
> by the spelling where historically closed syllables like -et, -ait,
> -ais have /ɛ/ and historically open ones like -é, -ai have /e/.

Suggested correction:
This is largely indicated
by the spelling where historically OPEN syllables like -et, -ait,
-ais have /ɛ/ and historically CLOSED ones like -é, -ai have /e/.

> This distinction is breaking down and "appears to be reliably
> maintained only in [...] frequent lexical items"; in particular the
> conditional and future tense endings -rais and -rai have effectively
> merged.
>
> For /œ/-/ø/ and /ɔ/-/o/ the situation is reversed: There is no /œ/
> or /ɔ/ in CV syllables. There is one regular modification for CVC
> syllables: final /z/ forces /o/ and /ø/. Apart from that, the
> opposition is somewhat maintained in CVC syllables. For /ø/ vs.
> /œ/ the examples are very strained: veule - veulent, jeûne - jeune.
> There are more potential minimal pairs for /o/ vs. /ɔ/, where /o/
> is frequently spelled -ô- or -au-.
>
> For non-final syllables the mid-open/mid-close distinction is
> neutralized. Here vowel harmony can be frequently observed where
> the mid vowels assimilate to the height of the final vowel.
>
> I think overall it is fairly clear that the system is in a state
> of collapse and that the formerly phonemic distinctions [e]-[ɛ],
> [ø]-[œ], [o]-[ɔ] are well into the process of becoming allophonic.
>
> ----
>
> Two further vowel mergers are worth mentioning:
>
> The dictionary opposition /a/-/ɑ/ is dead. As recent editions of
> Grevisse point out exasperatedly, even among the speakers who
> maintain the distinction, there is too little agreement on which
> word has which vowel.
>
> The two nasal vowels /ɛ̃/ and /œ̃/ have merged in Parisian French.
> Notice for instance the widespread cutesy abbreviation "1fo" for
> "info". This seems to have crept up on everybody largely unnoticed,
> since I think far less ink has been spilled over it than over the
> other vowel changes.
> --
> Christian "naddy" Weisgerber na...@mips.inka.de

On Saturday, June 5, 2021 at 8:30:06 AM UTC-7, Christian Weisgerber wrote:
> On 2021-06-04, António Marques <anton...@sapo.pt> wrote:
>
> > In the french I was taught, <ai> was just the same as <è>. I've since been
> > made aware that is often the same as <é>, but not the rules for that. This
> > looks like as good as time as any other to inquire about those.
> > What are the rules? Quelles sont-elles, les règles?
> That is part of the bigger question how the mid-close /e, ø, o/ and
> mid-open vowels /ɛ, œ, ɔ/ are distributed. The answer to that is
> made complicated by regional differences--don't trust the Belgian!--and
> the fact that the descriptive rules from 1900 have fossilized into
> prescriptive ones.
>
> Based on Fagyal/Kibbee/Jenkins, _French: A Linguistic Introduction_,
> I'll try to summarize for Northern Metropolitan (= Parisian) French.
>
> The basic rule is that CVC syllables have /ɛ, œ, ɔ/ and CV syllables
> have /e, ø, o/. There are significant exceptions in word-final
> syllables:
>
> There is no /e/ in CVC syllables, but in CV syllables, an opposition
> between /e/ and /ɛ/ is somewhat maintained. This is largely indicated
> by the spelling where historically closed syllables like -et, -ait,
> -ais have /ɛ/ and historically open ones like -é, -ai have /e/.

How about a closed syllable spelled with <é>? I've heard [depAtma] from
an Anglophone affecting French pronunciation. Would French listeners
flag his [e] as wrong?

> This distinction is breaking down and "appears to be reliably
> maintained only in [...] frequent lexical items"; in particular the
> conditional and future tense endings -rais and -rai have effectively
> merged.
>
> For /œ/-/ø/ and /ɔ/-/o/ the situation is reversed: There is no /œ/
> or /ɔ/ in CV syllables.

<fleur> ends with [œ] to my ear. If the r was pronounced, I didn't
notice it.

> There is one regular modification for CVC
> syllables: final /z/ forces /o/ and /ø/.

In the CVC context <com>/ <comme>, the pronunciation is described
as [kɔm] but I find the [ɔ] quality hardly perceptile; they sound rather like
English <come>. In <come> in a Sheffield accent is where I hear [kɔm];
I've also heard John Lennon pronounce <stunned> as [stɔnd].
Confusingly, in an English dictionary's pronunciation guide, [kɔm] is
not that regional pronunciation of <come> but a non-rhotic
pronunciation of <corm> which I'd prefer to see as [kɔ:m]..

>. Apart from that, the
> opposition is somewhat maintained in CVC syllables. For /ø/ vs.
> /œ/ the examples are very strained: veule - veulent, jeûne - jeune.
> There are more potential minimal pairs for /o/ vs. /ɔ/, where /o/
> is frequently spelled -ô- or -au-.
>
> For non-final syllables the mid-open/mid-close distinction is
> neutralized. Here vowel harmony can be frequently observed where
> the mid vowels assimilate to the height of the final vowel.
>
> I think overall it is fairly clear that the system is in a state
> of collapse and that the formerly phonemic distinctions [e]-[ɛ],
> [ø]-[œ], [o]-[ɔ] are well into the process of becoming allophonic.
> ----
>
> Two further vowel mergers are worth mentioning:
>
> The dictionary opposition /a/-/ɑ/ is dead. As recent editions of
> Grevisse point out exasperatedly, even among the speakers who
> maintain the distinction, there is too little agreement on which
> word has which vowel.
>
> The two nasal vowels /ɛ̃/ and /œ̃/ have merged in Parisian French.
> Notice for instance the widespread cutesy abbreviation "1fo" for
> "info". This seems to have crept up on everybody largely unnoticed,
> since I think far less ink has been spilled over it than over the
> other vowel changes.
> --
> Christian "naddy" Weisgerber na...@mips.inka.de

Sat, 5 Jun 2021 07:01:11 -0700 (PDT): "Peter T. Daniels"
<grammatim@verizon.net> scribeva:

>On Saturday, June 5, 2021 at 9:40:04 AM UTC-4, Ruud Harmsen wrote:
>> Fri, 4 Jun 2021 12:35:51 -0700 (PDT): "Peter T. Daniels"
>> <gram...@verizon.net> scribeva:
>> >On Friday, June 4, 2021 at 3:16:02 PM UTC-4, wugi wrote:
>> >> Op 4/06/2021 om 19:58 schreef Peter T. Daniels:
>> >
>> >DO NOT DELETE ATTRIBUTIONS
>> >
>> >> >> Not surprising. The phonetic difference between [e] and [I] is largely
>> >> >> undefined.
>> >> > What an odd thing to say. If those two phones belong to different
>> >> I agree with Ruud.
>> >Maybe he is misusing "undefined." What do you think he meant by it?
>>
>> A clear discription of what the difference is, in acoustic,
>> articulary, auditory terms. Height, positition on front-back scale,
>> formants.
>
>Perhaps in your autodidacticism you've never encountered the prolific
>authors Peter Ladefoged and Ian Maddieson.

Ladefoged is a VERY well known name to me, Maddieson is not.

>They provide charts
>galore of the formant-based vowel triangles of English and many other
>languages averaged from dozens (or more; fewer for endangered languages,
>of course) of speakers.

I am talking about IPA in general, not about specific languages.

>> >> > phonemes in one's native language, one has no trouble distinguishing
>> >> > them. (Anyway you probably meant [E] and [I]?)
>> >> No.
>> >[e] and [I] are quite different
>> >> > For [e]/[I], mate/mitt in English; for [E]/[I], met/mitt. long in meer.
>> >> For [e]/[I], smeer/smeed/smid in Dutch. [I] (very) short in smid, [e.]
>> >> half long in smeed, [e:] long in smeer. Same mouth position each vowel.
>> >If those are phonemically different, there is no reason for Ruud
>> >to find them hard to distinguish.
>>
>> Smeer and smeed aren't phonemically different, only phonetically (for
>> me, probably not for Wugi), conditioned by syllable-final /r/.
>
>Oy. Sounds like the British r-less problem --

Historically related, yes. Many forms of English, Dutch, German,
Danish have vocalisations of syllable-final r, although the details
vary. They affect the vowels.

>they have to posit a whole
>extra series of surface phonemes to account for the vowels before
>former /r/.

We don't. Not yet. Allophones.

--
Ruud Harmsen, http://rudhar.com

Sat, 5 Jun 2021 06:49:15 -0700 (PDT): "Peter T. Daniels"
<grammatim@verizon.net> scribeva:

>On Saturday, June 5, 2021 at 7:54:35 AM UTC-4, Dingbat wrote:
>> On Friday, June 4, 2021 at 10:58:42 AM UTC-7, Peter T. Daniels wrote:
>> > On Friday, June 4, 2021 at 11:54:33 AM UTC-4, Ruud Harmsen wrote:
>> > > Fri, 4 Jun 2021 01:07:28 -0700 (PDT): Dingbat
>> > > <ranjit_...@yahoo.com> scribeva:
>
>> > > >I find this word enigmatic:
>> > > >encastré
>> > > >https://forvo.com/word/encastr%C3%A9/
>> > > >The terminal vowel sounds like [I]!
>> > > > (the ending vowel of ancestry)
>> > (In early-20th-century RP. I don't know where Indian English stands
>> > on that suffix.)
>> > > Not surprising. The phonetic difference between [e] and [I] is largely
>> > > undefined.
>> > What an odd thing to say. If those two phones belong to different
>> > phonemes in one's native language, one has no trouble distinguishing
>> > them. (Anyway you probably meant [E] and [I]?)
>>
>> He couldn't have meant [E]. It is too far from [I] to be identified as [I].
>
>Sorry, but in the American English dialect of St. Louis, Missouri (and
>environs), "pin" and "pen" have merged. [I] and [E] are similar enough
>to have become indistinguishable before [n]. (People there write with
>an inkpen. Chicagoans find that amusing.)
>
>> > For [e]/[I], mate/mitt in English; for [E]/[I], met/mitt.
>> > > I clearly hear a cardinal [e], however, as expected.
>>
>> mate has [eI], not [e].
>
>Depends on your phonemic analysis. British phoneticians prefer
>to write [e:].

Do they? Never ever seen that.

--
Ruud Harmsen, http://rudhar.com

Sat, 5 Jun 2021 05:33:05 -0700 (PDT): Arnaud Fournet
<fournet.arnaud@wanadoo.fr> scribeva:

>Le vendredi 4 juin 2021 à 14:21:15 UTC+2, wugi a écrit :
>> There aren't clear ones, and if there are, they're not universal. From
>> what there is, the distinction is better maintained in eg. Belgium
>> (except dialectical) than in 'mainstream' Île de France (which is also
>> in the stage of coinciding -in and -un).
>>
>> -ai ~ é: parlai ~ parlé, parler, parlez etc.
>> -ais, -ait, -aient ~ è: mais ~ mets
>> -aise ~ è: (ma)laise ~ (mé)lèze.
>
>It's the theoretical phonology of Parisian French,

of the 1950s or before. But it seems a lot of shifts and mergers have
been gradually taking place in the meantime.

>but I seriously doubt present-day speakers abide by these laws.

Right.

Isn't 'parlai' a largely theoretical and litterary form anyway? If
people don't use a form in day-to-day speech, they no longer learn it
as a child, so how should they know how it's pronounced?
>> But the increased confusion é,è shows up in spelling errors between
>> parlai and parlais, parlerai and parlerais (besides between homophonic
>> parlé, parler and parlez). Also in spelling confusion éléverai/élèverai;
>> accéderai/accèderai....
>
>There's a lot of confusion in actual speech and idiolects.

--
Ruud Harmsen, http://rudhar.com

Sat, 5 Jun 2021 05:28:09 -0700 (PDT): Arnaud Fournet
<fournet.arnaud@wanadoo.fr> scribeva:

>Le vendredi 4 juin 2021 à 11:55:29 UTC+2, António Marques a écrit :
>> Athel Cornish-Bowden <acor...@imm.cnrs.fr> wrote:
>> > On 2021-06-04 08:33:21 +0000, mabel wugi said:
>> >
>> >> Op 4/06/2021 om 10:28 schreef mabel wugi:
>> >>> Op 4/06/2021 om 10:07 schreef Dingbat:
>> >>>
>> >>>>
>> >>>> I find this word enigmatic:
>> >>>> encastré
>> >>>> https://forvo.com/word/encastr%C3%A9/
>> >>>>
>> >>
>> >>>
>> >>>> The <c> is pronounced like in Italian!
>> >>>
>> >>> Regular.
>> >>
>> >>
>> >> Have a look at this doublet: châtrer - castrer
>> >> https://www.larousse.fr/dictionnaires/francais/châtrer/14931
>> >> https://www.larousse.fr/dictionnaires/francais/castrer/13677
>> >
>> > Not to be confused with Castrais, which refers to someone from Castres.
>> > We know someone from Castres, and when we first met him and we asked
>> > him if he was from this region (Provence) he said "Non, je suis
>> > Castrais". As intended, we understood Castrais as castré until he
>> > explained.
>> >
>> >
>> In the french I was taught, <ai> was just the same as <è>. I've since been -,
>> made aware that is often the same as <é>, but not the rules for that. This
>> looks like as good as time as any other to inquire about those.
>> What are the rules? Quelles sont-elles, les règles?
>
>In theory, as is inherited from Middle French,
>digraph -ai# should be closed é
>digraph -ai- followed by -e(s), -s, -t, -ent, should be open è
>Now, it's obvious that present-day speakers tend to harmonize their oral practice to spelling,

Why is that obvious?

>so that most people who have a distinction between é and è would pronounce ai as è in all cases,
>which from a diachronical point of view is incorrect.

>In theory, inherited phonology would mean aimai is émé, but aimais is émè,

Not èmé and èmè? We were taught (in 1966) to pronounce the more common
words aimer, aimez, aimé and aimée all as èmé.

>gai is gé but gaie is gè, etc.
>I seriously doubt that present-day francophones respect that.

>About all have spelling-driven phonology, not inherited phonology.

What about inherited but shifted, changed by more recent developments?
People learn such things are young children, long before they can read
and write.

Our 2 and 3 year olds can both perfectly pronounce the infamous Dutch
<ui>, which all foreign later learners struggle with, and they never
confuse <ei/ij> with <ee>, which our Argentinian queen keeps finding
difficult, despite her intensive lessons and now very good Dutch.

I remember a scene probably over a year ago, in which I showed the now
3,5 and egg and an onion, and asked here if she recognised and could
repeat the sound difference, <ei> and <ui>. She looked at me with a
face expressing "how on earth could you think I would NOT know that?
It's too easy."

>Of course, all those who have no distinction between é and è are unaware of these issues.

--
Ruud Harmsen, http://rudhar.com

Sat, 5 Jun 2021 16:36:57 -0700 (PDT): Dingbat
<ranjit_mathews@yahoo.com> scribeva:

>On Saturday, June 5, 2021 at 8:30:06 AM UTC-7, Christian Weisgerber wrote:
>> On 2021-06-04, António Marques <anton...@sapo.pt> wrote:
>>
>> > In the french I was taught, <ai> was just the same as <è>. I've since been
>> > made aware that is often the same as <é>, but not the rules for that. This
>> > looks like as good as time as any other to inquire about those.
>> > What are the rules? Quelles sont-elles, les règles?
>> That is part of the bigger question how the mid-close /e, ø, o/ and
>> mid-open vowels /?, œ, ?/ are distributed. The answer to that is
>> made complicated by regional differences--don't trust the Belgian!--and
>> the fact that the descriptive rules from 1900 have fossilized into
>> prescriptive ones.
>>
>> Based on Fagyal/Kibbee/Jenkins, _French: A Linguistic Introduction_,
>> I'll try to summarize for Northern Metropolitan (= Parisian) French.
>>
>> The basic rule is that CVC syllables have /?, œ, ?/ and CV syllables
>> have /e, ø, o/. There are significant exceptions in word-final
>> syllables:
>>
>> There is no /e/ in CVC syllables, but in CV syllables,
>
>I've heard an Anglophone pronounce DEPARTMENT as
>[depAtma]. Would the French find the [e] wrong?
>
>> an opposition
>> between /e/ and /?/ is somewhat maintained.
>> This is largely indicated
>> by the spelling where historically closed syllables like -et, -ait,
>> -ais have /?/ and historically open ones like -é, -ai have /e/.
>
>Suggested correction:
>This is largely indicated
>by the spelling where historically OPEN syllables like -et, -ait,
>-ais have /?/ and historically CLOSED ones like -é, -ai have /e/.

Closed syllables are CVC, open syllables are CV.
In many languages, like Dutch and French, open syllables tend to have
close vowels, and closed syllables have open vowels. Confusing.

--
Ruud Harmsen, http://rudhar.com

Sat, 5 Jun 2021 19:52:17 -0700 (PDT): Dingbat
<ranjit_mathews@yahoo.com> scribeva:
>How about a closed syllable spelled with <é>? I've heard [depAtma] from
> an Anglophone affecting French pronunciation. Would French listeners
> flag his [e] as wrong?

It isn't an open syllable. Départment is dé-part-ment, the syllables
are open-closed-closed, the vowels close (high), neutral, back
nasalised, [depaRtmA~].

>> This distinction is breaking down and "appears to be reliably
>> maintained only in [...] frequent lexical items"; in particular the
>> conditional and future tense endings -rais and -rai have effectively
>> merged.
>>
>> For /œ/-/ø/ and /?/-/o/ the situation is reversed: There is no /œ/
>> or /?/ in CV syllables.
>
><fleur> ends with [œ] to my ear. If the r was pronounced, I didn't
> notice it.

It is pronounced, but involves some vocalisation.

--
Ruud Harmsen, http://rudhar.com

On Sunday, June 6, 2021 at 4:52:19 AM UTC+2, Dingbat wrote:
> In the CVC context <com>/ <comme>, the pronunciation is described
> as [kɔm] but I find the [ɔ] quality hardly perceptile; they sound rather like
> English <come>.

<com> would produce a nasalised vowel. <comme> does not.
The /ɔ/ tends to lose its rounding in modern French.
https://fr.wikipedia.org/wiki/Prononciation_du_fran%C3%A7ais#cite_note-8
"La voyelle /ɔ/ est souvent « désarrondie » et se rapproche du /ʌ/."

> In <come> in a Sheffield accent is where I hear [kɔm];
> I've also heard John Lennon pronounce <stunned> as [stɔnd].
> Confusingly, in an English dictionary's pronunciation guide, [kɔm] is
> not that regional pronunciation of <come> but a non-rhotic
> pronunciation of <corm> which I'd prefer to see as [kɔ:m]..

That Northern English vowel is an unrounded [o], so that's [ɤ] phonetically.
It occurs for historic short /u/, regardless of whether that has shifted in
Standard (i.e. Southern) English. The words <but> and <put> have the same
vowel in such pronunciation styles (like they used to have everywhere). Also
also <looked>, <foot>, etc. <Look> and <luck> are homophones there.

> Confusingly, in an English dictionary's pronunciation guide, [kɔm] is
> not that regional pronunciation of <come> but a non-rhotic
> pronunciation of <corm> which I'd prefer to see as [kɔ:m]..

Yes, me too. And always seen as such.

Sat, 5 Jun 2021 16:14:01 -0700 (PDT): Dingbat
<ranjit_mathews@yahoo.com> scribeva:

>On Saturday, June 5, 2021 at 7:30:06 AM UTC-7, Christian Weisgerber wrote:
>> On 2021-06-05, Dingbat <ranjit_...@yahoo.com> wrote:
>>
>> > I don't remember an English chart but English doesn't have an [e].
>> At that level of detail you need to specify which variant of English.
>> Southern hemisphere accents have raised the DRESS vowel to [e].
>
>DRESS has [e] in Malayalis' English also but I wasn't counting such accents.
>The only thing I remember from an Oz English chart is that /aI/ is realized as [EI].

More like [OI]?
--
Ruud Harmsen, http://rudhar.com

Le dimanche 6 juin 2021 à 07:42:54 UTC+2, Ruud Harmsen a écrit :
> Sat, 5 Jun 2021 05:28:09 -0700 (PDT): Arnaud Fournet
> <fournet...@wanadoo.fr> scribeva:
> >Le vendredi 4 juin 2021 à 11:55:29 UTC+2, António Marques a écrit :
> >> Athel Cornish-Bowden <acor...@imm.cnrs.fr> wrote:
> >> > On 2021-06-04 08:33:21 +0000, mabel wugi said:
> >> >
> >> >> Op 4/06/2021 om 10:28 schreef mabel wugi:
> >> >>> Op 4/06/2021 om 10:07 schreef Dingbat:
> >> >>>
> >> >>>>
> >> >>>> I find this word enigmatic:
> >> >>>> encastré
> >> >>>> https://forvo.com/word/encastr%C3%A9/
> >> >>>>
> >> >>
> >> >>>
> >> >>>> The <c> is pronounced like in Italian!
> >> >>>
> >> >>> Regular.
> >> >>
> >> >>
> >> >> Have a look at this doublet: châtrer - castrer
> >> >> https://www.larousse.fr/dictionnaires/francais/châtrer/14931
> >> >> https://www.larousse.fr/dictionnaires/francais/castrer/13677
> >> >
> >> > Not to be confused with Castrais, which refers to someone from Castres.
> >> > We know someone from Castres, and when we first met him and we asked
> >> > him if he was from this region (Provence) he said "Non, je suis
> >> > Castrais". As intended, we understood Castrais as castré until he
> >> > explained.
> >> >
> >> >
> >> In the french I was taught, <ai> was just the same as <è>. I've since been -,
> >> made aware that is often the same as <é>, but not the rules for that. This
> >> looks like as good as time as any other to inquire about those.
> >> What are the rules? Quelles sont-elles, les règles?
> >
> >In theory, as is inherited from Middle French,
> >digraph -ai# should be closed é
> >digraph -ai- followed by -e(s), -s, -t, -ent, should be open è
> >Now, it's obvious that present-day speakers tend to harmonize their oral practice to spelling,
> Why is that obvious?
> >so that most people who have a distinction between é and è would pronounce ai as è in all cases,
> >which from a diachronical point of view is incorrect.
>
> >In theory, inherited phonology would mean aimai is émé, but aimais is émè,
> Not èmé and èmè? We were taught (in 1966) to pronounce the more common
> words aimer, aimez, aimé and aimée all as èmé.

This sounds quite incorrect, both synchronically and diachronically.

On Saturday, June 5, 2021 at 11:09:02 PM UTC-7, Ruud Harmsen via Google Groups <google@rudhar.com> wrote:
> On Sunday, June 6, 2021 at 4:52:19 AM UTC+2, Dingbat wrote:
> > In the CVC context <com>/ <comme>, the pronunciation is described
> > as [kɔm] but I find the [ɔ] quality hardly perceptile; they sound rather like
> > English <come>.
> <com> would produce a nasalised vowel. <comme> does not.
> The /ɔ/ tends to lose its rounding in modern French.
> https://fr.wikipedia.org/wiki/Prononciation_du_fran%C3%A7ais#cite_note-8
> "La voyelle /ɔ/ est souvent « désarrondie » et se rapproche du /ʌ/."
> > In <come> in a Sheffield accent is where I hear [kɔm];
> > I've also heard John Lennon pronounce <stunned> as [stɔnd].
> > Confusingly, in an English dictionary's pronunciation guide, [kɔm] is
> > not that regional pronunciation of <come> but a non-rhotic
> > pronunciation of <corm> which I'd prefer to see as [kɔ:m]..

> That Northern English vowel is an unrounded [o], so that's [ɤ] phonetically.

I've been called LOVE pronounced as [lɤv] by a waitress near Hull. Weird
way to address a stranger, I thought. So, there's at least one word with [ɤ].
But I'm sure that I've heard SLUG as not [slɤg] but [slɔg], in Manchester.

> It occurs for historic short /u/, regardless of whether that has shifted in
> Standard (i.e. Southern) English. The words <but> and <put> have the same
> vowel in such pronunciation styles (like they used to have everywhere). Also
> also <looked>, <foot>, etc. <Look> and <luck> are homophones there.

I remember a punny joke about a WW pilot's story about being chased by
Fokkers being heard as Fuckers. There is a surname Focker with too
open a vowel to be thus heard.

> > Confusingly, in an English dictionary's pronunciation guide, [kɔm] is
> > not that regional pronunciation of <come> but a non-rhotic
> > pronunciation of <corm> which I'd prefer to see as [kɔ:m].

> Yes, me too. And always seen as such.

On Sunday, June 6, 2021 at 1:10:58 AM UTC-4, Ruud Harmsen wrote:
> Sat, 5 Jun 2021 07:01:11 -0700 (PDT): "Peter T. Daniels"
> <gram...@verizon.net> scribeva:
> >On Saturday, June 5, 2021 at 9:40:04 AM UTC-4, Ruud Harmsen wrote:
> >> Fri, 4 Jun 2021 12:35:51 -0700 (PDT): "Peter T. Daniels"
> >> <gram...@verizon.net> scribeva:
> >> >On Friday, June 4, 2021 at 3:16:02 PM UTC-4, wugi wrote:
> >> >> Op 4/06/2021 om 19:58 schreef Peter T. Daniels:

> >> >DO NOT DELETE ATTRIBUTIONS

> >> >> >> Not surprising. The phonetic difference between [e] and [I] is largely
> >> >> >> undefined.
> >> >> > What an odd thing to say. If those two phones belong to different
> >> >> I agree with Ruud.
> >> >Maybe he is misusing "undefined." What do you think he meant by it?
> >> A clear discription of what the difference is, in acoustic,
> >> articulary, auditory terms. Height, positition on front-back scale,
> >> formants.
> >Perhaps in your autodidacticism you've never encountered the prolific
> >authors Peter Ladefoged and Ian Maddieson.
>
> Ladefoged is a VERY well known name to me, Maddieson is not.
>
> >They provide charts
> >galore of the formant-based vowel triangles of English and many other
> >languages averaged from dozens (or more; fewer for endangered languages,
> >of course) of speakers.
>
> I am talking about IPA in general, not about specific languages.

Utterly meaningless. The Internat Phon Assoc has tried to provide enough
symbols that they can be used to write all _phonemic_ distinctions in any
language. (For a few rare contrasts, they resort to diacritics, such as for the
few languages that contrast dental and alveolar points of articulation.)

Daniel Jones's "Cardinal Vowels" system is orthogonal to IPA. It can only
\be learned by personal instruction direct from Jones, or from those who
had it from him (Ladefoged was one of his very last students) -- but he
made audio recordings of the Cardinal Vowels several times during his
career, and the several recordings _do not agree_ among themselves.

Maddieson is Ladefoged's most prominent student -- now retired --
and is the co-author of the most important phonetics reference guide
there is (Ladefoged & Maddieson 1996). (Australian-origin; pronounced
like Madison.)

> >> >> > phonemes in one's native language, one has no trouble distinguishing
> >> >> > them. (Anyway you probably meant [E] and [I]?)
> >> >> No.
> >> >[e] and [I] are quite different
> >> >> > For [e]/[I], mate/mitt in English; for [E]/[I], met/mitt. long in meer.
> >> >> For [e]/[I], smeer/smeed/smid in Dutch. [I] (very) short in smid, [e.]
> >> >> half long in smeed, [e:] long in smeer. Same mouth position each vowel.
> >> >If those are phonemically different, there is no reason for Ruud
> >> >to find them hard to distinguish.
> >> Smeer and smeed aren't phonemically different, only phonetically (for
> >> me, probably not for Wugi), conditioned by syllable-final /r/.
> >Oy. Sounds like the British r-less problem --
>
> Historically related, yes. Many forms of English, Dutch, German,
> Danish have vocalisations of syllable-final r, although the details
> vary. They affect the vowels.
>
> >they have to posit a whole
> >extra series of surface phonemes to account for the vowels before
> >former /r/.
>
> We don't. Not yet. Allophones.

It would be simpler to just not drop the r's/