Glottal stops in English speech (was: Re: Intro C: Mini-FAQ )
Igor Merfert
Neil Coffey wrote:
> [...]
> The English word 'hour' begins with a vowel. [1]
> [...]
> [1] Exception: when stressed at the beginning of an utterance, it
> may begin with a glottal stop.
Once I was told that English does not use glottal stops to begin sounds
with and that this fact would be the reason why a glottal stop is always
a dead give-away for English speaking Germans.
On the other hand the pronunciation expert R. Easton said London Cockney
uses lots of glottal stops.
What is right?
... or is it a question of the glottal stop severity that makes it a
give-away? And if yes, how can it be determined? Could it be possible
that it depends on which sound is preceded by a glottal stop? And
finally, are there glottal-stop-free dialects in English?
Cheers,
Igor
NG switched from AUE to sci.lang
Harlan Messinger
Igor Merfert <mer...@et.uni-magdeburg.de> wrote:
: Neil Coffey wrote:
:> [...]
:> The English word 'hour' begins with a vowel. [1]
:> [...]
:> [1] Exception: when stressed at the beginning of an utterance, it
:> may begin with a glottal stop.
: Once I was told that English does not use glottal stops to begin sounds
: with and that this fact would be the reason why a glottal stop is always
: a dead give-away for English speaking Germans.
I commonly have a glottal stop at the beginning of an utterance that would
otherwise start with a vowel. For example, I have one if I read the
preceding sentence out loud.
Within a sentence, a glottal stop may be prefixed to a word starting with
a vowel if the delivery is emphatic or drawn out. Normally, if I say "the
answer to the question is" or "I would like some answers", there is no
glottal stop before "answers". But if I were speaking very deliberately,
one might go in. Germans, I believe, would normally put a glottal stop in
that place, unless they knew to do otherwise or had unconsciously picked
up the usual lack of one in English.
Alwyn Thomas
Igor Merfert wrote:
> On the other hand the pronunciation expert R. Easton said London Cockney
> uses lots of glottal stops.
An awful lot [lo?], mainly as an allophone of /t/. This pronunciation is
now widespread in many parts of England and Scotland.
Alwyn
Antonio
Neil Coffey wrote:
> The English word 'hour' begins with a vowel.
Would I be correct to presume that no etymological h- ever existed, rather
was only a spelling innovation?
--
am - read 'untalnew', not 'entonio'
What will it Death advance thy name/Upon cold rocks to waste a flame
Or by mistake it seems to throw/Light torches into pits of snow?
Your rage is lost by a killing frost/And with your arrows you may try
To make the harmed or aged bleed/But indeed not compel one heart to die!
Alwyn Thomas
Antonio wrote:
> Neil Coffey wrote:
>
> > The English word 'hour' begins with a vowel.
>
> Would I be correct to presume that no etymological h- ever existed, rather
> was only a spelling innovation?
"Hour" comes from Old French *hore*, in which the <h> was certainly
"etymological". The Latins, on the other hand, must at some stage have
pronounced the <h> in *hora*, otherwise it wouldn't be there.
Alwyn
Harlan Messinger
Antonio <ip20...@ip.pt> wrote:
>Neil Coffey wrote:
>
>> The English word 'hour' begins with a vowel.
>
>Would I be correct to presume that no etymological h- ever existed, rather
>was only a spelling innovation?
It's always been there, going all the way back to Greek "hora" =
"season", as in "horoscope". In Middle English, "ure" or "our" did
appear as variants, according to Partridge.
Pete Carlton
>
> On the other hand the pronunciation expert R. Easton said London Cockney
> uses lots of glottal stops.
My (west coast) English often has a glottal stop for final /t/... anyone
else notice this?
pete carlton
MCB Dept., UC Berkeley
345 LSA Berkeley CA 94704-3200,
510-643-8277
pe...@candelab.berkeley.edu
pcar...@uclink4.berkeley.edu
http://www.ocf.berkeley.edu:80/~pcarlton
Antonio
Alwyn Thomas wrote:
>>> The English word 'hour' begins with a vowel.
>>
>> Would I be correct to presume that no etymological h- ever existed,
>> rather was only a spelling innovation?
>
> "Hour" comes from Old French *hore*, in which the <h> was certainly
> "etymological". The Latins, on the other hand, must at some stage have
> pronounced the <h> in *hora*, otherwise it wouldn't be there.
Yes, but what about german _Uhr_?
I mean, I'd think that these ancient loans had already entered english
without h-; why else would 'real' h- words that have h- still nowadays be
different? A multitude of reasons, I suppose...
I thought OF was 'huere', but I know little of the issue anyway :(
Bram M. & Christina C.
I've been analyzing /t/ in my dialect for some time now (I was raised in
Las Vegas, NV), and it has a number of peculiarities, including all
final /t/ = glottal stop. I also assimilate /t/ with a preceeding /n/,
i.e. I do not distinguish "inner-net" and "internet". Also, /t/ before
syllabic /n/ is also a glottal for me, for instance the t's in "button"
are a glottal stop rather than a flap or usual /t/. I get laughed at
sometimes now in Seattle though, so maybe it's just a California/Nevada
thing.
Bram
Alwyn Thomas
Antonio wrote:
> Alwyn Thomas wrote:
> > "Hour" comes from Old French *hore*, in which the <h> was certainly
> > "etymological". The Latins, on the other hand, must at some stage have
> > pronounced the <h> in *hora*, otherwise it wouldn't be there.
>
> Yes, but what about german _Uhr_?
According to Wahrig, German *Uhr* is from Italian *ora*, Old French
*(h)ore*, *eure*.
> I mean, I'd think that these ancient loans had already entered english
> without h-; why else would 'real' h- words that have h- still nowadays be
> different? A multitude of reasons, I suppose...
Neither German *Uhr* nor English "hour" are as ancient as you seem to think.
There was a time when h was never pronounced in English; I think its
reintroduction was a triumph of eighteenth-century prescriptivism.
> I thought OF was 'huere', but I know little of the issue anyway :(
There are many differenct spellings. <hore> is what I found in my English
dictionary.
Alwyn
Tom Wier
Alwyn Thomas wrote:
> > I mean, I'd think that these ancient loans had already entered english
> > without h-; why else would 'real' h- words that have h- still nowadays be
> > different? A multitude of reasons, I suppose...
>
> Neither German *Uhr* nor English "hour" are as ancient as you seem to think.
>
> There was a time when h was never pronounced in English; I think its
> reintroduction was a triumph of eighteenth-century prescriptivism.
*Never* pronounced? Surely you mean in Southern England
or in the standard language. There must have been other
dialects (like near Scotland or something) that preserved
the /h/, otherwise I find it difficult to believe that
the prescriptivists could have been so successful (witness
their failures like "split infinitives", sentence
final prepositions, or double negatives, all of which
have _long_ histories in some community or another).
Though it's true many dialects lack the /h/ to this day...
=======================================
Tom Wier <senorb...@mail.utexas.edu>
ICQ#: 4315704 AIM: Deuterotom
Website: <http://www.angelfire.com/tx/eclectorium/>
"Cogito ergo sum, sed credo ergo ero."
"Sunny day, chasing the clouds away" translates
in Klingon as "Day of the daytime star, the
clouds are filled with dread and forced to flee".
=======================================
Tom Wier
Bram M. & Christina C. wrote:
> I've been analyzing /t/ in my dialect for some time now (I was raised in
> Las Vegas, NV), and it has a number of peculiarities, including all
> final /t/ = glottal stop.
I thought I did the same thing for a while, but then I realized
that for me it's more of an unreleased /t/, rather than a glottal
stop. But I wouldn't be at all surprised to find that glottal stops
in that position are general among Americans.
> I also assimilate /t/ with a preceeding /n/,
> i.e. I do not distinguish "inner-net" and "internet". Also, /t/ before
> syllabic /n/ is also a glottal for me, for instance the t's in "button"
> are a glottal stop rather than a flap or usual /t/. I get laughed at
> sometimes now in Seattle though, so maybe it's just a California/Nevada
> thing.
No, I think you're right. Here in Texas, we do the same thing,
and I've never heard anything different. In fact, if someone
did use [b@tn,] rather than [b@?n,], I would probably look at
them funny (it would probably sound more like pretensiousness
than a legitimate dialect).
Antonio
Alwyn Thomas wrote:
> Neither German *Uhr* nor English "hour" are as ancient as you seem to
> think.
But it is a useful word.
My Duden gives no cognate of _Stunde_ for the British Isles after Old
English.
> There was a time when h was never pronounced in English; I think its
> reintroduction was a triumph of eighteenth-century prescriptivism.
Most interesting! I didn't know it.
Alwyn Thomas
Antonio wrote:
> My Duden gives no cognate of _Stunde_ for the British Isles after Old
> English.
I've often wondered where Welsh *awr* ("hour") comes from. Hardly from
Latin *ho^ra*, since then you would have **w^r*.
Alwyn
Alwyn Thomas
Tom Wier wrote:
> *Never* pronounced? Surely you mean in Southern England
> or in the standard language. There must have been other
> dialects (like near Scotland or something) that preserved
> the /h/, otherwise I find it difficult to believe that
> the prescriptivists could have been so successful (witness
> their failures like "split infinitives", sentence
> final prepositions, or double negatives, all of which
> have _long_ histories in some community or another).
Most English dialects still drop the h.
Another case where the prescriptivists triumphed is in restoring the final [N]
sound in words like "nothing' and "talking".
Alwyn
Antonio
Alwyn Thomas wrote:
> Most English dialects still drop the h.
I suppose so, but Tom is probably correct at least in what concerns Scots?
> Another case where the prescriptivists triumphed is in restoring the
> final [N] sound in words like "nothing' and "talking".
Hae' they?
--
am - read 'untalnew', not 'entonio'
What will it Death advance thy name/Upon cold rocks to waste a flame
Or by mistake it seems to throw/Light torches into pits of snow?
Your rage is lost by a killin frost/And with your arrows you may try
Alwyn Thomas
Antonio wrote:
> Alwyn Thomas wrote:
>
> > Most English dialects still drop the h.
>
> I suppose so, but Tom is probably correct at least in what concerns Scots?
I would imagine so, but then I'm no expert.
Alwyn
Alwyn Thomas
Antonio wrote:
> My Duden gives no cognate of _Stunde_ for the British Isles after Old
> English.
I think "stound" made it into Modern English, just as *stond" made it into
Modern Dutch. Neither is current today, however.
Alwyn
Tom Wier
Alwyn Thomas wrote:
>
> Tom Wier wrote:
>
> > *Never* pronounced? Surely you mean in Southern England
> > or in the standard language. There must have been other
> > dialects (like near Scotland or something) that preserved
> > the /h/, otherwise I find it difficult to believe that
> > the prescriptivists could have been so successful (witness
> > their failures like "split infinitives", sentence
> > final prepositions, or double negatives, all of which
> > have _long_ histories in some community or another).
>
Really? That's surprising. I personally have never heard any,
but then I've not travelled much in the UK.
> Another case where the prescriptivists triumphed is in restoring the final [N]
> sound in words like "nothing' and "talking".
Yeah, I could see how that could have been reintroduced.
(As an aside:
In my historical linguistics class last semester, one of
the examples was on different speakers' perceptions of
how often they "drop the g" in [-IN] suffixes. Something
like only 8% of the speakers correctly guessed how often
they used syllabic /-n/ for /-IN/ (with most of course
vastly overestimating how often they used /-IN/). This
might say something about how often most people think about
language in general, though they use it all the time.)
I could easily see how after the sound law that changed
[N] to phonemic status (by literally dropping the g, [Ng])
occurred, the resulting /N/ could move forward back to /n/.
=======================================
Tom Wier <tw...@ccwf.cc.utexas.edu>
ICQ#: 4315704 AIM: Deuterotom
Website: <http://www.angelfire.com/tx/eclectorium/>
"Cogito ergo sum, sed credo ergo ero."
=======================================
Colin Fine
In article <6mmis9$9...@bgtnsc03.worldnet.att.net>, Tom Wier <Tomaso.Hous
t...@postoffice.worldnet.att.net> writes
>Bram M. & Christina C. wrote:
>
>
>No, I think you're right. Here in Texas, we do the same thing,
>and I've never heard anything different. In fact, if someone
>did use [b@tn,] rather than [b@?n,], I would probably look at
>them funny (it would probably sound more like pretensiousness
>than a legitimate dialect).
>
rather than any variety of American. I think the most common realisation
of this stop is post-velar release through the nasal cavity, the tongue
retaining alveolar closure.
--
-----------------------------------------------------------------------
| Colin Fine 66 High Ash, Shipley, W Yorks. BD18 1NE, UK |
| Tel: 01274 592696/0976 635354 e-mail: co...@kindness.demon.co.uk |
| "Don't just do something! Stand there!" |
| - from 'Behold the Spirit' (workshop) |
-----------------------------------------------------------------------
Tom Wier
Colin Fine wrote:
> >No, I think you're right. Here in Texas, we do the same thing,
> >and I've never heard anything different. In fact, if someone
> >did use [b@tn,] rather than [b@?n,], I would probably look at
> >them funny (it would probably sound more like pretensiousness
> >than a legitimate dialect).
> >
> Are you sure you use a glottal stop there? It sounds Cockney to me
> rather than any variety of American. I think the most common realisation
> of this stop is post-velar release through the nasal cavity, the tongue
> retaining alveolar closure.
Well, here the whole case is a little confused by the fact that the
following nasal is itself alveolar, which might lead one to believe
that it's really /b@tn/, where /t/ is realized as an unreleased
stop, rather than the other possible allophone, [?].
At first I kinda thought you were right, but then I realized it
that my closure only came when I was producing the following
alveolar nasal /n/.
But you're right though: glottal stops are not something that's
just definitive of an American accent.
=======================================
Tom Wier <tw...@ccwf.cc.utexas.edu>
ICQ#: 4315704 AIM: Deuterotom
Website: <http://www.angelfire.com/tx/eclectorium/>
"Cogito ergo sum, sed credo ergo ero."
God is subtle, but he is not malicious. -A. Einstein
=======================================
Tom Wier
Greg Lee wrote:
> Tom Wier <Tomaso....@postoffice.worldnet.att.net> wrote:
>
> : (As an aside:
> : In my historical linguistics class last semester, one of
> : the examples was on different speakers' perceptions of
> : how often they "drop the g" in [-IN] suffixes. Something
> : like only 8% of the speakers correctly guessed how often
> : they used syllabic /-n/ for /-IN/ (with most of course
> : vastly overestimating how often they used /-IN/). This
> : might say something about how often most people think about
> : language in general, though they use it all the time.)
>
> : I could easily see how after the sound law that changed
> : [N] to phonemic status (by literally dropping the g, [Ng])
> : occurred, the resulting /N/ could move forward back to /n/.
>
> At one time, presumably, the phonemic form of the -ing suffix
> was /Ing/, pronounced [INg]. If the /g/ is lost, /In/
> is left, which can become [schwa n] or syllable [n].
Here's where I don't follow you. Do you mean the lossof /g/ results in
(the phonetic form) [IN] or [In]? I would
think that the loss of the /g/ would only cause the [N] to
become a phonemic difference, one noted by the speakers
(to become /IN/, rather than /In/).
So, it'd go like this:
Phonetic Phonemic
Early stage: -[Ing] -/Ing/
Nasal Assimilation: -[INg] -/Ing/
drop of /g/: -[IN] -/IN/
(phonemicization occurs here; later, under "casual" conditions:
[N] -> [n] / I _ # and where stress does not fall on that
syllable (thus leaving such words as "sing", "bring", etc., intact),
but on this see below.)
Thus, the loss of the /g/ brings about the phonemicization
of [N] to /N/.
> If the [g] is lost, though, the vowel of [IN] does not
> reduce to schwa, but under the influence of the following
> [N] is a high central vowel.
Ummm... well, not in my dialect. In mine, it's /@/ (or syllabic /n/
or whatever), or, if there's spelling pronunciation going on, then
it might be /I/ (a high front lax vowel), but not a high
central vowel (though /I/ is certainly more central than /i/).
Do you mean /I/ by "high central vowel"?
(I'm not quite sure how you got from your comment above
to the one below.)
> The [N] may then, especially
> in casual speech, shift to an [n]. So you can tell whether
> the final [n] is from /n/ or from /N/ (perhaps /ng/) by
> whether the preceding vowel reduces to schwa or remains [I],
> on the one hand, or is a high central vowel.
Well, that would be true, except that in my dialect, the form
denoted by the "lack of a g", as in "playin'", is phonetically
[plei@n] (or if you wish, [plein,], where [,] denotes syllabicity),
*with* the schwa.
(It is of course possible that my dialect has generalized a
a normal distinction between /I/ and /@/ (i.e., that final words
that would otherwise end in -/In/ now end in -/n,/), but I have never
heard any other dialects that distinctly pronounced /I/ in
such a location. It is also probable that the normal rule of
vowel reduction in unstressed syllables is coming into play
here.)
> I'm not sure this is a correct analysis for the pronunciation
> variants [In], [schwa n], [high-central-i n], but it does
> suggest that it is useful to mind the distinction between
> phonemes and allophones in such an historical analysis.
Oh, Of course, no disagreein' with ya there. That's always
something that's important to consider.
Speaking of vowels before nasals, has anyone
every noticed the tense vowel /i/ rather than /I/? I think I
myself use the former rather than the latter, even though
"Chomsky doesn't say so" (not like that matters, though).
If I do, I think the two're in free variation (pace Labov, et al.).
=======================================
Tom Wier <tw...@ccwf.cc.utexas.edu>
ICQ#: 4315704 AIM: Deuterotom
Website: <http://www.angelfire.com/tx/eclectorium/>
"Cogito ergo sum, sed credo ergo ero."
=======================================
Greg Lee
Tom Wier <Tomaso....@postoffice.worldnet.att.net> wrote:
: (As an aside:
: In my historical linguistics class last semester, one of
: the examples was on different speakers' perceptions of
: how often they "drop the g" in [-IN] suffixes. Something
: like only 8% of the speakers correctly guessed how often
: they used syllabic /-n/ for /-IN/ (with most of course
: vastly overestimating how often they used /-IN/). This
: might say something about how often most people think about
: language in general, though they use it all the time.)
: I could easily see how after the sound law that changed
: [N] to phonemic status (by literally dropping the g, [Ng])
: occurred, the resulting /N/ could move forward back to /n/.
At one time, presumably, the phonemic form of the -ing suffix
was /Ing/, pronounced [INg]. If the /g/ is lost, /In/
is left, which can become [schwa n] or syllable [n].
If the [g] is lost, though, the vowel of [IN] does not
reduce to schwa, but under the influence of the following
[N] is a high central vowel. The [N] may then, especially
in casual speech, shift to an [n]. So you can tell whether
the final [n] is from /n/ or from /N/ (perhaps /ng/) by
whether the preceding vowel reduces to schwa or remains [I],
on the one hand, or is a high central vowel.
I'm not sure this is a correct analysis for the pronunciation
variants [In], [schwa n], [high-central-i n], but it does
suggest that it is useful to mind the distinction between
phonemes and allophones in such an historical analysis.
--
Greg Lee <l...@Hawaii.edu>
PG Carter
Tom Wier wrote:
>
>
> So, it'd go like this:
> Phonetic Phonemic
> Early stage: -[Ing] -/Ing/
> Nasal Assimilation: -[INg] -/Ing/
> drop of /g/: -[IN] -/IN/
>
> (phonemicization occurs here; later, under "casual" conditions:
> [N] -> [n] / I _ # and where stress does not fall on that
> syllable (thus leaving such words as "sing", "bring", etc., intact),
> but on this see below.)
I'm not an expert in this area, but a couple of points occur to me that
might be worth throwing into the arena. Firstly, there are varieties of
English (eg much of the Midlands and NW of England) which still use
[INg]. It's quite a prestigious form in Liverpool, for example.
Secondly, on the [n]/[N] variation in words like "walking" which doesn't
crop up in words like "sing" or "bring": English inflectional morphology
works with apicals, so [-s/z] for plurals or 3s pres verb ending, [-n]
for many older plurals, [-t/d] for past endings. The odd one out is [N]
(which is velar -- dorsal rather than apical), so perhaps from a
stuctural point of view it's not surprising that the "-ing" words
usually appear with [-n] rather than [-N]. Of course, those "-ing"
words which never have [-n] are precisely those where the "-ing" is not
morphological, so there's no structural reason to bar non-apicals.
Paul
--
_ Paul Carter ____________________________________ pgc...@york.ac.uk _
Dept of Language & Linguistic Science | http://www.york.ac.uk/~pgc104/
University of York | tel: +44 (0)1904 432660
Heslington, York. YO10 5DD | fax: +44 (0)1904 432673
Greg Lee
Tom Wier <senorb...@mail.utexas.edu> wrote:
: Greg Lee wrote:
...
: > At one time, presumably, the phonemic form of the -ing suffix
: > was /Ing/, pronounced [INg]. If the /g/ is lost, /In/
: > is left, which can become [schwa n] or syllable [n].
: Here's where I don't follow you. Do you mean the lossof /g/ results in
: (the phonetic form) [IN] or [In]? I would
: think that the loss of the /g/ would only cause the [N] to
: become a phonemic difference, one noted by the speakers
: (to become /IN/, rather than /In/).
I'm distinguishing between the loss of /g/ and the loss of [g].
Take /g/ away from /Ing/ -- that leaves /In/, whose /n/ cannot
become [N], since there's no [g] for it to assimilate to.
I am suggesting that this is the source of the pronunciation
variant [@n]. This explains why the unstressed [I] reduces to [@],
since this reduction does not occur before [N].
However, if [g] is lost, take away the [g] from [bar-i Ng]
leaves [bar-i N]. (Some might have, or transcribe, [i] where
I have [bar-i].) The [bar-i] is not subject to reduction.
There is a further change that makes this pattern opaque.
The [N] -> [n] in the -ing ending, but this leaves the
preceding vowel still unreduced. So, e.g., for me,
"Bering" or "ball-bearing" always end in [N], but
the participle "bearing" can be either [bEr bar-i N]
or [bEr bar-i n], but never (for this particular verb)
*[bEr@n], which would mean "baron", or [bErIn], which
would be the proper name "Berin".
I don't to imply that loss-of-[g] and loss-of-/g/ are
sound changes or are synchronic rules, necessarily. I
don't know about that.
: So, it'd go like this:
: Phonetic Phonemic
: Early stage: -[Ing] -/Ing/
: Nasal Assimilation: -[INg] -/Ing/
: drop of /g/: -[IN] -/IN/
Here, you must mean "drop of [g]", unless you think there was
a preceding phonemicizarion of /Ing/ to /INg/.
: (phonemicization occurs here; later, under "casual" conditions:
: [N] -> [n] / I _ # and where stress does not fall on that
: syllable (thus leaving such words as "sing", "bring", etc., intact),
: but on this see below.)
I don't think so. I don't think unstressed [I] before [N] is possible.
I'm sure I don't say it, and I don't think I've ever heard it.
: Thus, the loss of the /g/ brings about the phonemicization
: of [N] to /N/.
It may be that /N/ has become a phoneme in contemporary English.
It's a disputed point.
...
: Well, that would be true, except that in my dialect, the form
: denoted by the "lack of a g", as in "playin'", is phonetically
: [plei@n] (or if you wish, [plein,], where [,] denotes syllabicity),
: *with* the schwa.
I have here either [pley@n] (/pleyIn/), [pleyIn] (/pleyIn/),
[pley bar-i N] (/pleyIN/), or [pley bar-i n] (/pleyIN/).
(Writing [ey] for the diphthong sometimes written [ei].)
...
--
Greg Lee <l...@Hawaii.edu>
D. Edward Gund v. Brighoff
In article <358F8AFA...@pia.bt.co.uk>,
Alwyn Thomas <al...@pia.bt.co.uk> wrote:
>Antonio wrote:
>
>> My Duden gives no cognate of _Stunde_ for the British Isles after Old
>> English.
>
>Latin *ho^ra*, since then you would have **w^r*.
Nevertheless, that's what the Geiriadur Prifysgol Cymru claims. I'd sus-
pect an English source except that the diphthongised form is also present
in Cornish and Breton.
--
Daniel "Da" von Brighoff /\ Dilettanten
(de...@midway.uchicago.edu) /__\ erhebt Euch
/____\ gegen die Kunst!
D. Edward Gund v. Brighoff
In article <3590DF9D...@pia.bt.co.uk>,
Alwyn Thomas <al...@pia.bt.co.uk> wrote:
>Antonio wrote:
>
>> My Duden gives no cognate of _Stunde_ for the British Isles after Old
>> English.
>
The OED backs you up on this. It may still survive in some dialects.
>just as *stond" made it into
>Modern Dutch. Neither is current today, however.
Alwyn Thomas
D. Edward Gund v. Brighoff wrote:
> In article <358F8AFA...@pia.bt.co.uk>,
> Alwyn Thomas <al...@pia.bt.co.uk> wrote:
> >I've often wondered where Welsh *awr* ("hour") comes from. Hardly from
> >Latin *ho^ra*, since then you would have **w^r*.
>
> Nevertheless, that's what the Geiriadur Prifysgol Cymru claims. I'd sus-
> pect an English source except that the diphthongised form is also present
> in Cornish and Breton.
Welsh derivation from Latin is very regular. Why this exception?
I wonder if one could postulate a borrowing from Old French - about whose
phonology I know very little, however. Has anyone any ideas?
Alwyn
Keith C. Ivey
Greg Lee <l...@Hawaii.edu> wrote:
>It may be that /N/ has become a phoneme in contemporary English.
>It's a disputed point.
For give my ignorance, but how do those who dispute the point analyze
the distinction between "longer" (/lONg@r/, the comparative of the
adjective "long") and "longer" (/lON@r/, the agent form of the verb
"long")? And why don't "finger" and "singer" rhyme in standard
dialects if they both have /ng/ in the same environment?
Keith C. Ivey <kci...@cpcug.org>
http://cpcug.org/user/kcivey/
Washington, DC
Peter T. Daniels
Keith C. Ivey wrote:
> For give my ignorance, but how do those who dispute the point analyze
> the distinction between "longer" (/lONg@r/, the comparative of the
> adjective "long") and "longer" (/lON@r/, the agent form of the verb
> "long")? And why don't "finger" and "singer" rhyme in standard
> dialects if they both have /ng/ in the same environment?
The environment isn't identical: there's a morpheme boundary in sing+er
but not in finger.
That of course doesn't help with longer and longer.
Some people in phonology class in 1970 (I think they were from the
Bronx) claimed they had a g in (clothes) hanger but not in (airplane)
hangar.
--
Peter T. Daniels gram...@worldnet.att.net
D. Edward Gund v. Brighoff
In article <3592D3B5...@dircon.co.uk>,
The Old French and Anglo-French form would have been <(h)(o)ure>, pro-
nounced /ur@/. This, like Latin (h)o:ra, should have yielded Modern Welsh
*ur. Why it didn't really is a puzzling question.
The only other source languages that come to mind are Old Irish and Old
Norse. (Modern Irish has 'uair' and Modern Icelandic has 'u/r', neither
of which is particularly helpful.)
Alwyn Thomas
D. Edward Gund v. Brighoff wrote:
> The Old French and Anglo-French form would have been <(h)(o)ure>, pro-
> nounced /ur@/. This, like Latin (h)o:ra, should have yielded Modern Welsh
> *ur. Why it didn't really is a puzzling question.
>
> The only other source languages that come to mind are Old Irish and Old
> Norse. (Modern Irish has 'uair' and Modern Icelandic has 'u/r', neither
> of which is particularly helpful.)
Then I have two further suggestions:
i) *awr* derives from a native Celtic word
ii) it derives from Latin horarium > *wrawr > *rawr > awr, the initial r being
lost by confusion with the definite article.
Alwyn
Miguel Carrasquer Vidal
On 29 Jun 98 04:03:15 GMT, lib...@jaka.ece.uiuc.edu (Dave Librik)
wrote:
>Morris-Jones says (Welsh Grammar, sec. 71.ii.3):
> [Note: I've ASCIIfied the phonetics as follows:
> [open-o] = reversed-c, Cardinal Vowel 6
> V: = long vowel, macron over V
> V, = cedilla under vowel
> V^ = circumflex, acen gron, Greek curved circumflex over vowel]
>
>"In _awr_ `hour' and _nawn_ `noon' we have _aw_ < Lat. _o:_. These
> have been explained as late borrowings; but historically this is
> improbable. Possibly the pronunciation of _ho:ra_ varied in Lat.,
> since Greek [omega] (= o,:) was popularly sounded _o,:_ (glo:^ssa
> [gamma][lambda][omega-with-circumflex][sigma][sigma][alpha] > Ital.
> chio,sa); _o:,_ would give [open-o] > _aw_. For _nawn_ see sec. 76
> iii (4)."
>
>The reason [open-o] goes to _aw_ is the regular development:
> Brit. a: > Early W. [open-o] > Old W. au (in final syll), o (in non-final).
> JMJ has a nice picture on p. 94 here which I will not attempt to draw...
>
>I have no way to judge the validity of this explanation; I don't even know
>what the little hook under the _o_ means in Italian. Maybe one of the
>Romance specialists could give an opinion here?
The custom among Romanists is to notate the open vowels by a little
hook (ogonek) under the vowel, and close vowels by a dot under.
The symbols are mainly used in combination with e and o, where <e,>
is IPA <E> (epsilon) and <o,> is IPA <O> (reversed c = open o).
Latin long /o:/ in principle develops into Romance close /o/ (<o.> in
Romanist notation), not open /O/ (<o,>).
==
Miguel Carrasquer Vidal ~ ~
Amsterdam _____________ ~ ~
m...@wxs.nl |_____________|||
========================== Ce .sig n'est pas une .cig
Alwyn Thomas
Dave Librik wrote:
> Morris-Jones says (Welsh Grammar, sec. 71.ii.3):
> [Note: I've ASCIIfied the phonetics as follows:
> [open-o] = reversed-c, Cardinal Vowel 6
> V: = long vowel, macron over V
> V, = cedilla under vowel
> V^ = circumflex, acen gron, Greek curved circumflex over vowel]
>
> "In _awr_ `hour' and _nawn_ `noon' we have _aw_ < Lat. _o:_. These
> have been explained as late borrowings; but historically this is
> improbable. Possibly the pronunciation of _ho:ra_ varied in Lat.,
> since Greek [omega] (= o,:) was popularly sounded _o,:_ (glo:^ssa
> [gamma][lambda][omega-with-circumflex][sigma][sigma][alpha] > Ital.
> chio,sa); _o:,_ would give [open-o] > _aw_. For _nawn_ see sec. 76
> iii (4)."
>
> The reason [open-o] goes to _aw_ is the regular development:
> Brit. a: > Early W. [open-o] > Old W. au (in final syll), o (in non-final).
> JMJ has a nice picture on p. 94 here which I will not attempt to draw...
>
> I have no way to judge the validity of this explanation; I don't even know
> what the little hook under the _o_ means in Italian. Maybe one of the
> Romance specialists could give an opinion here?
Sir John Morris Jones was an outstanding scholar and man of letters, who did
much for the Welsh language. Hoewever, I do not agree with him on this occasion.
There is no evidence that I know of that *hora* was pronounced with a short or
open o in Latin or proto-romance. It is true that Italian *chiosa* ("gloss",
"commentary") does have an open o (that is what the hook means!), but that is
probably due to the contamination of Greek *glo^ssa* with Latin *clausula*.
Alwyn
Antonio
Alwyn Thomas wrote:
> There is no evidence that I know of that *hora* was pronounced with a
> short or open o in Latin or proto-romance. It is true that Italian
> *chiosa* ("gloss", "commentary") does have an open o (that is what
> the hook means!), but that is probably due to the contamination of
> Greek *glo^ssa* with Latin *clausula*.
But isn't the au supposed to turn into o. rather than o,?
Why is the basis for [AW] being written <ou> in English today?
German has <au> (sour/sauer); has the english pronounciation at any time
been different? That may be relevant to the development of _awr_.
--
am - read 'untalnew', not 'entonio'
What will it Death advance thy name/Upon cold rocks to waste a flame
Or by mistake it seems to throw/Light torches into pits of snow?
Your rage is lost by a killing frost/And with your arrows you may try
Alwyn Thomas
Antonio wrote:
> But isn't the au supposed to turn into o. rather than o,?
The outcome is variable: clavus > chiOdo ("nail"), audio > Odo ("I hear"),
aurum > Oro ("gold"), causa > cOsa, but cauda > coda ("tail").
> Why is the basis for [AW] being written <ou> in English today?
> German has <au> (sour/sauer); has the english pronounciation at any time
> been different?
My understanding is, something like: u: > ow > Ow > @w > aw
> That may be relevant to the development of _awr_.
I don't see how.
Alwyn
J. A. Rea
In article <35976f07....@news.wxs.nl>
m...@wxs.nl (Miguel Carrasquer Vidal) writes:
>
>Latin long /o:/ in principle develops into Romance close /o/ (<o.> in
>Romanist notation), not open /O/ (<o,>).
>
vowels simplyl coalesce, so that /o/ and /o:/ both end up being /o/, leaving
a simple five vowel system ( /aj/ and the few remaining /oj/ coalescing
also with /e/ and /e:/ into simple /e/, while /aw/ usually yields /a/.
John
Ki semenat ispinaza, non andet iskultsu!
J. A. Rea ja...@ukcc.uky.edu
P. S. When you say "Romance", I of course presume you mean Sard.
Miguel Carrasquer Vidal
On Mon, 29 Jun 98 20:00:49 EDT, JA...@ukcc.uky.edu (J. A. Rea) wrote:
>In article <35976f07....@news.wxs.nl>
>m...@wxs.nl (Miguel Carrasquer Vidal) writes:
>
>>
>>Latin long /o:/ in principle develops into Romance close /o/ (<o.> in
>>Romanist notation), not open /O/ (<o,>).
>>
>Define "Romance", please.
Allright, I meant 7-vowel Romance. Sard and Romanian don't play.
==
Miguel Carrasquer Vidal ~ ~
Amsterdam _____________ ~ ~
m...@wxs.nl |_____________|||
========================== Ce .sig n'est pas une .cig
Peter T. Daniels
Miguel Carrasquer Vidal wrote:
>
> On Mon, 29 Jun 98 20:00:49 EDT, JA...@ukcc.uky.edu (J. A. Rea) wrote:
>
> >In article <35976f07....@news.wxs.nl>
> >m...@wxs.nl (Miguel Carrasquer Vidal) writes:
> >
> >>
> >>Latin long /o:/ in principle develops into Romance close /o/ (<o.> in
> >>Romanist notation), not open /O/ (<o,>).
> >>
> >Define "Romance", please.
>
> Allright, I meant 7-vowel Romance. Sard and Romanian don't play.
In a few weeks, Carl Winter Verlag (Heidelberg) will publish *The
Origins of the Romance Languages* by Giuliano Bonfante (written ca. 1940
at Princeton in English; edited and being published by his daughter the
Etruscologist Larissa Bonfante of NYU in time for his 94th birthday in
August; I gave it a brief edit just before the final copy was produced).
It is essentially an elaboration and justification of Groeber's theory
(from the 1880s) that each Romance language preserves/represents the
state of the Latin language at the time the area where it is spoken was
settled. Thus Sardinian is the most archaic Romance language, and
Italian is the most advanced--since it never "hived off" and started
undergoing independent developments. In 150 or so pages, he collects a
stupendous amount of data showing correlations between the Romance
languages and the successive stages of Latin.
The publisher promises that the price will be reasonable (I think I saw
something like DM50, ca. $35), and I'm eager to find out what Romanists
will think of it.
Is this a legitimate POV? Is it taken for granted in Romance linguistics
so that this publication is superfluous? Or, has the idea been
definitively refuted and rejected? Or, has attention not been paid?
The new Foreword by Edward Tuttle of UCLA praises the breadth of the
author's knowledge and the wealth of dialect data brought to bear; but
of course it's so flowery and gracious one can't tell if it's actually
saying whether the book is worth reading.
Will everyone be rushing out for a copy, or is it already a best-ignored
museum piece?
STAN MULAIK
"Peter T. Daniels" <gram...@worldnet.att.net> writes:
>In a few weeks, Carl Winter Verlag (Heidelberg) will publish *The
>Origins of the Romance Languages* by Giuliano Bonfante (written ca. 1940
>at Princeton in English; edited and being published by his daughter the
>Etruscologist Larissa Bonfante of NYU in time for his 94th birthday in
>August; I gave it a brief edit just before the final copy was produced).
The theory on which this account is based is a bit dated; but is it still
plausible? What other accounts of the origins of the Romance Languages
would you recommend? I've read Robert Hall's popular version of this,
but did not find the detail I'd like compared, say, to Bourciez's 1946
book _Elements de Linguistique Romane_. I'd like to know if there is
anything more recent than these?
--
Stanley A. Mulaik
School of Psychology, Georgia Institute of Technology, Atlanta, GA 30332
uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!pscccsm
Internet: psc...@prism.gatech.edu
Peter T. Daniels
STAN MULAIK wrote:
>
> "Peter T. Daniels" <gram...@worldnet.att.net> writes:
>
> >In a few weeks, Carl Winter Verlag (Heidelberg) will publish *The
> >Origins of the Romance Languages* by Giuliano Bonfante (written ca. 1940
> >at Princeton in English; edited and being published by his daughter the
> >Etruscologist Larissa Bonfante of NYU in time for his 94th birthday in
> >August; I gave it a brief edit just before the final copy was produced).
>
> The theory on which this account is based is a bit dated; but is it still
> plausible? What other accounts of the origins of the Romance Languages
> would you recommend? I've read Robert Hall's popular version of this,
> but did not find the detail I'd like compared, say, to Bourciez's 1946
> book _Elements de Linguistique Romane_. I'd like to know if there is
> anything more recent than these?
Hall, *External History of the Romance Languages* (the first of three
published volumes of a projected 8 or 10 on all aspects of Romance
linguistics; Phonology and Morphology also came out, and I wonder what
is to be found in his Nachlass). I don't know of a popular treatment by
him.
Most recent is Rebecca Posner, in the Cambridge Language Surveys; also
the Routledge volume edited by Harris & Vincent. Older is Elcock, *The
Romance Languages* in the Faber & Faber series. It's what Hall used in
his class in 1971 or so (he also gave out the Ext. Hist. as he finished
typing and dittoing it--this was before Xeroxing was widespread).
But I'm not qualified to recommend; we should hear from John Rea and
Miguel C V.
Miguel Carrasquer Vidal
>In a few weeks, Carl Winter Verlag (Heidelberg) will publish *The
>Origins of the Romance Languages* by Giuliano Bonfante (written ca. 1940
>at Princeton in English; edited and being published by his daughter the
>Etruscologist Larissa Bonfante of NYU in time for his 94th birthday in
>August; I gave it a brief edit just before the final copy was produced).
>It is essentially an elaboration and justification of Groeber's theory
>(from the 1880s) that each Romance language preserves/represents the
>state of the Latin language at the time the area where it is spoken was
>settled. Thus Sardinian is the most archaic Romance language, and
>Italian is the most advanced--since it never "hived off" and started
>undergoing independent developments. In 150 or so pages, he collects a
>stupendous amount of data showing correlations between the Romance
>languages and the successive stages of Latin.
The real story is probably more complicated than that. It usually
is. But certainly the Latin as spoken by the first Roman settlers in
each of the areas is one aspect of the whole complex story. For DM
50, I'd certainly consider looking at Bonfante's book.
mign...@my-dejanews.com
In article <35981AA...@dircon.co.uk>,
al...@dircon.co.uk wrote:
>
> Antonio wrote:
>
> > But isn't the au supposed to turn into o. rather than o,?
>
> The outcome is variable: clavus > chiOdo ("nail"), audio > Odo ("I hear"),
> aurum > Oro ("gold"), causa > cOsa, but cauda > coda ("tail").
Latin co:da is also attested (in M. Terentius Varro) along with cauda. I
thought that all Latin au's > Italian O's in stressed syllables of
non-learned words (i.e., those that aren’t direct loans from Latin).
Exceptions would be however welcome!
Michele Ignelzi
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/rg_mkgrp.xp Create Your Own Free Member Forum
Alwyn Thomas
mign...@my-dejanews.com wrote:
> Latin co:da is also attested (in M. Terentius Varro) along with cauda.
Thank you. I didn't know that.
> I
> thought that all Latin au's > Italian O's in stressed syllables of
> non-learned words (i.e., those that aren’t direct loans from Latin).
> Exceptions would be however welcome!
Well, there's *foce* from *fauces* - not a common word, however. Might it be
learned?
Alwyn
mign...@ats.it
In article <359E44CE...@dircon.co.uk>,
al...@dircon.co.uk wrote:
>
> mign...@my-dejanews.com wrote:
>
> > I
> > thought that all Latin au's > Italian O's in stressed syllables of
> > Exceptions would be however welcome!
>
> be
> learned?
Probably not, since learned "loans" normally retain *au*: from *fauces*
(usually only plural and ablative sing.) we have *fauci* (only pl., learned)
and *foce* (non-learned and rather common, I daresay). :) You hit the mark! I
can only be a little suspicious on account of the defective character of the
word...
Michele
Erland Sommarskog
kci...@cpcug.org (Keith C. Ivey) skriver:
>Greg Lee <l...@Hawaii.edu> wrote:
>>It may be that /N/ has become a phoneme in contemporary English.
>>It's a disputed point.
>
>the distinction between "longer" (/lONg@r/, the comparative of the
>adjective "long") and "longer" (/lON@r/, the agent form of the verb
>"long")? And why don't "finger" and "singer" rhyme in standard
>dialects if they both have /ng/ in the same environment?
And forgive an even more ignorant, but how do these people look at
"sin" and "sing"?
--
Erland Sommarskog, Stockholm, som...@algonet.se
F=F6r =F6vrigt anser jag att QP b=F6r f=F6rst=F6ras.
B=65sid=65s, I think QP should b=65 d=65stroy=65d.
Keith C. Ivey
>>For give my ignorance, but how do those who dispute the point analyze
>>the distinction between "longer" (/lONg@r/, the comparative of the
>>adjective "long") and "longer" (/lON@r/, the agent form of the verb
>>"long")? And why don't "finger" and "singer" rhyme in standard
>>dialects if they both have /ng/ in the same environment?
>
>And forgive an even more ignorant, but how do these people look at
>"sin" and "sing"?
That's a little easier. I assume they transcribe them as something
like /sIn/ and /sIng/ (realized as [sIN]).
J. A. Rea
In article <6nbss4$i...@bgtnsc03.worldnet.att.net>
"Peter T. Daniels" <gram...@worldnet.att.net> writes:
>STAN MULAIK wrote:
>>
>> "Peter T. Daniels" <gram...@worldnet.att.net> writes:
>>
>> >Origins of the Romance Languages* by Giuliano Bonfante (written ca. 1940
>>
>> plausible? What other accounts of the origins of the Romance Languages
>> would you recommend? I've read Robert Hall's popular version of this,
>
>Miguel C V.
>--
Bob Hall took strong issue with Bonfante. See his Introductory Linguistics.
A few years back there was a mud slinging contest between M. L. Wagner and
Walter von Wartburg as to whether the "archaic" character of Sard was due
to the early Latin colonization of the Island, or to its early isolation
from the rest of the Romania. Them Krauts can really sling the verbiage.
As Omar Fitzgerald said, (I paraphrase from memory) "but (I) evermore came
out the same door where in I went." I once decided to see if the then
newly devised Glottochronology would provide an answer the the Sard part of
the problem, and ended up with a noisy series of articles against the
method.
As for all of the above learned gentlemen, requiescant.... (another bloody
subjunctive!) The enterprise of classifying Romance languages should
be taken to the law courts and declared an attractive nuisance. (Don't
comment unless you understand the legal term " " .
Pax
Peter T. Daniels
J. A. Rea wrote:
>
> In article <6nbss4$i...@bgtnsc03.worldnet.att.net>
> "Peter T. Daniels" <gram...@worldnet.att.net> writes:
>
> >STAN MULAIK wrote:
> >>
> >> "Peter T. Daniels" <gram...@worldnet.att.net> writes:
> >>
> >> >In a few weeks, Carl Winter Verlag (Heidelberg) will publish *The
> >> >Origins of the Romance Languages* by Giuliano Bonfante (written ca. 1940
> >> > in time for his 94th birthday in
> >>
> >> The theory on which this account is based is a bit dated; but is it still
> >> plausible? What other accounts of the origins of the Romance Languages
> >> would you recommend? I've read Robert Hall's popular version of this,
> >
> >But I'm not qualified to recommend; we should hear from John Rea and
> >Miguel C V.
> >--
>
> Somebody rang my bell: nemo me impune clamat!
>
> Bob Hall took strong issue with Bonfante. See his Introductory Linguistics.
Larissa claims it was I who told her that Hall and Bonfante kissed and
made up, and then witin a few weeks were again slinging the mud--a
colleague asked one of them why, and he said it wasn't any fun not
having the worthy opponent to fight with.
I, on the other hand, am sure I first heard the story from her.
OTOH, it was definitely I who told her (and thus Giuliano) of the
existence of Hall's several memoirs, beginning with *Stormy Petrel in
Linguistics*, which were considerably more rude than what he put in his
more scholarly writing! She says G.B. loved the limericks written
against him.
>
> A few years back there was a mud slinging contest between M. L. Wagner and
> Walter von Wartburg as to whether the "archaic" character of Sard was due
> to the early Latin colonization of the Island, or to its early isolation
> from the rest of the Romania. Them Krauts can really sling the verbiage.
> As Omar Fitzgerald said, (I paraphrase from memory) "but (I) evermore came
> out the same door where in I went." I once decided to see if the then
> newly devised Glottochronology would provide an answer the the Sard part of
> the problem, and ended up with a noisy series of articles against the
> method.
Presumably this postdates Bonfante's book; it hasn't been revised to
take into account the 60 subsequent years of scholarship.
>
> As for all of the above learned gentlemen, requiescant.... (another bloody
> subjunctive!) The enterprise of classifying Romance languages should
> be taken to the law courts and declared an attractive nuisance. (Don't
> comment unless you understand the legal term " " .
>
> Pax
>
> Ki semenat ispinaza, non andet iskultsu!
>
> J. A. Rea ja...@ukcc.uky.edu
>
--
Peter T. Daniels gram...@worldnet.att.net
Antonio
>> and *foce* (non-learned and rather common, I daresay). :)
>
> But rather specialised: "mouth of a river", *metter foce in*, "to flow
> into".There are the only meanings I know.
Wow! This has nothing to do with the subject, but _foz_ [fo,S] is the
portuguese word for "mouth of a river", even entering some toponyms but
always in distinct form, while _foice_ [fo.ys] with archaic _fouce_
'sickle' is what I thought was the outcome of lat. _fauce_. Oi/ou is the
portuguese outcome of lat au, with oi being much less common and losing
ground in the standard speech, in the words where both are possible
(ouro/oiro 'gold', loic,a/louc,a 'geschirr', mouro/moiro 'moor',
loura/loira 'blonde', where the first is the more common form, but always
pouco 'paucus', louco 'crazy', etc). In words where both forms are
frequent, a [o@] compromise can be heard.
Anyway, what this means is that 1) either 'foz' is a loanword, which I
find very hard - though it would fit with the italian -, or 2) _fauce_ was
adopted independently at two times - which I also don't find probable -,
or 4) it mysteriously evolved into two very different words, or 5)
portuguese _foz_ and italian _foce_ don't come from _fauce at all, or
maybe they developed from a specialised variant of it, which would account
for the oddity of it in both languages. Notice, of course, that there was
a template in portuguese for the o, in _foz_ (o, is the 'open' vowel, I
hope I'm not mistaken) in regular words like voz < vox and noz < nux. Come
to think of it, the u in nux should've evolved into o. ... But there
aren't any o.z words in portuguese.
--
am - read 'untawnew', not 'entonio'
Come, let's watch the World / Put her Shape with rage
At this mass of Fools / Who mock the Holy Stage
And the Hand shall come down / To tear up the last page,
The Final Act / Of the Final Age!
Greg Lee
: >>It may be that /N/ has become a phoneme in contemporary English.
: >>It's a disputed point.
: >For give my ignorance, but how do those who dispute the point analyze
: >the distinction between "longer" (/lONg@r/, the comparative of the
: >adjective "long") and "longer" (/lON@r/, the agent form of the verb
: >"long")? And why don't "finger" and "singer" rhyme in standard
: >dialects if they both have /ng/ in the same environment?
: And forgive an even more ignorant, but how do these people look at
: "sin" and "sing"?
"longer" noun /longer/
"longer" adj /longre/
"sin" /sin/
"sing" /sing/
: --
: Erland Sommarskog, Stockholm, som...@algonet.se
: F=F6r =F6vrigt anser jag att QP b=F6r f=F6rst=F6ras.
: B=65sid=65s, I think QP should b=65 d=65stroy=65d.
--
Greg Lee <l...@Hawaii.edu>
Alwyn Thomas
> Alwyn Thomas wrote:
>
> >> and *foce* (non-learned and rather common, I daresay). :)
> >
> > But rather specialised: "mouth of a river", *metter foce in*, "to flow
> > into".There are the only meanings I know.
>
> Wow! This has nothing to do with the subject, but _foz_ [fo,S] is the
> portuguese word for "mouth of a river", even entering some toponyms but
> always in distinct form, while _foice_ [fo.ys] with archaic _fouce_
> 'sickle' is what I thought was the outcome of lat. _fauce_.
falcem > *fauce > fouce > foice
<snip>
> Anyway, what this means is that 1) either 'foz' is a loanword, which I
> find very hard - though it would fit with the italian -, or 2) _fauce_ was
> adopted independently at two times - which I also don't find probable -,
> or 4) it mysteriously evolved into two very different words, or 5)
> portuguese _foz_ and italian _foce_ don't come from _fauce at all, or
> maybe they developed from a specialised variant of it, which would account
> for the oddity of it in both languages. Notice, of course, that there was
> a template in portuguese for the o, in _foz_ (o, is the 'open' vowel, I
> hope I'm not mistaken) in regular words like voz < vox and noz < nux. Come
> to think of it, the u in nux should've evolved into o. ... But there
> aren't any o.z words in portuguese.
The plot thickens...
Alwyn
Colin Fine
<JA...@ukcc.uky.edu> writes
I am confused. When I read
>
>Somebody rang my bell: nemo me impune clamat!
>
I thought 'surely that should be subjunctive?'. I checked mentally -
clamo, clamare ... no, that's indicative. So what is the original?
'lacessit'. That sounds like an imperfect subjunctive (though wouldn't
that be 'lacesset'?). Don't know. Haven't got a Latin grammar here.
Then a bit later I saw .
>
>As for all of the above learned gentlemen, requiescant.... (another bloody
>subjunctive!)
which seems to suggest that 'clamat' is subjunctive after all.
Will somebody give me a clew out of the maze?
--
-----------------------------------------------------------------------
| Colin Fine 66 High Ash, Shipley, W Yorks. BD18 1NE, UK |
| Tel: 01274 592696/0976 635354 e-mail: co...@kindness.demon.co.uk |
| "Don't just do something! Stand there!" |
| - from 'Behold the Spirit' (workshop) |
-----------------------------------------------------------------------
Antonio
> Come to think of it, the u in nux should've evolved into o. ... But
> there aren't any o.z words in portuguese.
Except, of course, for _arroz_ 'rice' [@Ro.S] and the floating Estremoz
town.
In brazilian, btw, 'closed' (at least) vowels before -s or -z develop a
diphtong, so _arroz_ [Axo.ys] merges with _dois_ [do.ys] 'two', which,
btw, is _dous_ in galician.
Antonio
> I am confused. When I read
>
> >Somebody rang my bell: nemo me impune clamat!
>
> I thought 'surely that should be subjunctive?'. I checked mentally -
> clamo, clamare ... no, that's indicative. So what is the original?
> 'lacessit'. That sounds like an imperfect subjunctive (though wouldn't
> that be 'lacesset'?). Don't know. Haven't got a Latin grammar here.
My unknowlege of Latin tells me it is 'Nobody me in impunity calls',
meaning 'I had to write something because my name had been alluded to',
with _clamat_ as a simple 3rd person singular present indicative tense.
> Then a bit later I saw .
>
>> As for all of the above learned gentlemen, requiescant.... (another
>> bloody subjunctive!)
>
> which seems to suggest that 'clamat' is subjunctive after all.
>
> Will somebody give me a clew out of the maze?
Sorry, did my best. _Requiescant_ seems to me as utterly subjunctive but
_clamat_ doesn't.
J. A. Rea
Antonio <ip20...@ip.pt> writes:
>Colin Fine wrote:
>
>> >Somebody rang my bell: nemo me impune clamat!
>> I thought 'surely that should be subjunctive?'. I checked mentally -
>> clamo, clamare ... no, that's indicative. So what is the original?
>> 'lacessit'. That sounds like an imperfect subjunctive (though wouldn't
>> that be 'lacesset'?). Don't know. Haven't got a Latin grammar here.
>
(first singular present, of course, being the standard latin lemma)
and the ending is -it, which we might, for this short -ere conjugation,
presume to be present or perfect. Since all on this list immediately
spot the "Homeric" intentionally stolen and garbled quote, even if
they have only one eye and no dictionary, there is no problem.
>> Then a bit later I saw .
>>> As for all of the above learned gentlemen, requiescant.... (another
>>> bloody subjunctive!)
>>
>> which seems to suggest that 'clamat' is subjunctive after all.
>> Will somebody give me a clew out of the maze?
>
>Sorry, did my best. _Requiescant_ seems to me as utterly subjunctive but
>_clamat_ doesn't.
>
many fail to spot what is and what is not a subjunctive these days.
Pity. That's part f the reason they strangle on my sig. Oh well.
Ki semenat ispinaza, non andet iskultsu!
J. A. Rea ja...@ukcc.uky.edu
Erland Sommarskog
>Erland Sommarskog <som...@algonet.se> wrote:
>: And forgive an even more ignorant, but how do these people look at
>: "sin" and "sing"?
>
>"longer" noun /longer/
>"longer" adj /longre/
>"sin" /sin/
>"sing" /sing/
I have to say that this seems highly artificial to me. Of course, if
you are analysing a dialect in which "sing" is in fact realized as
[siNg] it makes sense, but most English dialets keep the <g> silent.
Is this some idea that new phonems can't arise? Hey, if /N/ can make
it's way into French and Swedish, why not English?
senorb...@mail.utexas.edu
som...@algonet.se (Erland Sommarskog) wrote:
>
> kci...@cpcug.org (Keith C. Ivey) skriver:
> >Greg Lee <l...@Hawaii.edu> wrote:
> >>It may be that /N/ has become a phoneme in contemporary English.
> >>It's a disputed point.
How could it not be? _People_hear_the_difference_ between
"sin" and "sing" (though see below). They wouldn't have it
in dictionaries if it were otherwise.
> >For give my ignorance, but how do those who dispute the point analyze
> >the distinction between "longer" (/lONg@r/, the comparative of the
> >adjective "long") and "longer" (/lON@r/, the agent form of the verb
> >"long")? And why don't "finger" and "singer" rhyme in standard
> >dialects if they both have /ng/ in the same environment?
>
> And forgive an even more ignorant, but how do these people look at
> "sin" and "sing"?
There is also, in my dialect, a slightly different vowel before
all velar nasals. My /I/ in such circumstances is much closer to
/i/ in many ways. But I don't know how far that runs, in general
American dialects (or British ones, for that matter).
=======================================
Tom Wier <senorb...@mail.utexas.edu>
ICQ#: 4315704 AIM: Deuterotom
Website: <http://www.angelfire.com/tx/eclectorium/>
"Cogito ergo sum, sed credo ergo ero."
Dies Irae, dies illa, | Quantus tremor est futurus!
Solvet saeclum in favilla,| Quando Judex est venturus!
Teste David cum Sibylla! | Cuncta stricte discussurus!
=======================================
senorb...@mail.utexas.edu
> >"longer" noun /longer/
> >"longer" adj /longre/
> >"sin" /sin/
> >"sing" /sing/
>
> I have to say that this seems highly artificial to me. Of course, if
> you are analysing a dialect in which "sing" is in fact realized as
> [siNg] it makes sense, but most English dialets keep the <g> silent.
>
> Is this some idea that new phonems can't arise? Hey, if /N/ can make
> it's way into French and Swedish, why not English?
It _has_ arisen. Go ask any man on the street if he hears a difference
between [sIn] and [sIN], without any context that might give away the meaning
and making sure those vowels are the same, and I'm sure you'll get some
response that they are indeed different sounds. This is all you need to do
show that it's a phoneme: find what might be a minimal pair and ask a native
speaker.
=======================================================
Tom Wier <senorb...@mail.utexas.edu>
ICQ#: 4315704 AIM: Deuterotom
Website: <http://www.angelfire.com/tx/eclectorium/>
"Cogito ergo sum, sed credo ergo ero."
Dies Irae, dies illa, | Quantus tremor est futurus!
Solvet saeclum in favilla,| Quando Judex est venturus!
Teste David cum Sibylla! | Cuncta stricte discussurus!
========================================================
Peter T. Daniels
>
> In article <6nouef$3ft$1...@zingo.tninet.se>,
> som...@algonet.se (Erland Sommarskog) wrote:
> >
> > kci...@cpcug.org (Keith C. Ivey) skriver:
> > >Greg Lee <l...@Hawaii.edu> wrote:
> > >>It may be that /N/ has become a phoneme in contemporary English.
> > >>It's a disputed point.
>
> How could it not be? _People_hear_the_difference_ between
> "sin" and "sing" (though see below). They wouldn't have it
> in dictionaries if it were otherwise.
Now you're running into definitions of "phoneme." According to
generative phonology (introduced by Morris Halle in *The Sound Pattern
of Russian*, 1959), there is no need for a level of analysis called
"phoneme"; and if, as in Chomsky and Halle, *The Sound Pattern of
English* (1968), you can get rid of some of the symbols (ex phonemes) in
the underlying forms by making them conditioned variants of other
symbols, even if they never surface, then you've achieved some precious
economy. Hence the previously presented distinction between |longer| and
|longre| (vertical brackets mark underlying forms in SPE).
SPE phonology was extremely influential for several years (the book was
recently reprinted in paperback, though I do wonder why), but it didn't
take too long for phonology to develop in different directions and
recover from its pernicious influence.
> There is also, in my dialect, a slightly different vowel before
> all velar nasals. My /I/ in such circumstances is much closer to
> /i/ in many ways. But I don't know how far that runs, in general
> American dialects (or British ones, for that matter).
I don't have that. Are you maybe hearing some sorts of diphthongs, as is
much more frequent in Texas than in various other varieties?
Peter T. Daniels
>
> In article <6obfrh$ph8$4...@zingo.tninet.se>,
> > >"longer" noun /longer/
> > >"longer" adj /longre/
> > >"sin" /sin/
> > >"sing" /sing/
> >
> > I have to say that this seems highly artificial to me. Of course, if
> > you are analysing a dialect in which "sing" is in fact realized as
> > [siNg] it makes sense, but most English dialets keep the <g> silent.
> >
> > Is this some idea that new phonems can't arise? Hey, if /N/ can make
> > it's way into French and Swedish, why not English?
>
> It _has_ arisen. Go ask any man on the street if he hears a difference
> between [sIn] and [sIN], without any context that might give away the meaning
> and making sure those vowels are the same, and I'm sure you'll get some
> response that they are indeed different sounds. This is all you need to do
> show that it's a phoneme: find what might be a minimal pair and ask a native
> speaker.
How charmingly pre- and post-Hallean!
Greg Lee
: In article <6obfrh$ph8$4...@zingo.tninet.se>,
: > >"longer" noun /longer/
: > >"longer" adj /longre/
: > >"sin" /sin/
: > >"sing" /sing/
: >
: > I have to say that this seems highly artificial to me. Of course, if
: > you are analysing a dialect in which "sing" is in fact realized as
: > [siNg] it makes sense, but most English dialets keep the <g> silent.
: >
: > Is this some idea that new phonems can't arise? Hey, if /N/ can make
: > it's way into French and Swedish, why not English?
: It _has_ arisen. Go ask any man on the street if he hears a difference
: between [sIn] and [sIN], without any context that might give away the meaning
: and making sure those vowels are the same, and I'm sure you'll get some
...
I certainly never said or implied that "sin" and "sing" might be
phonemically the same. If you will just look above, you'll see
two different phonemic forms for them, but neither with /N/.
: response that they are indeed different sounds. This is all you need to do
: show that it's a phoneme: find what might be a minimal pair and ask a native
: speaker.
Even if it were a minimal pair, that doesn't prove the differing
segments are different phonemes. In the /N/-less analysis, they
derive from different phoneme sequences.
--
Greg Lee <l...@Hawaii.edu>
senorb...@mail.utexas.edu
Greg Lee <l...@Hawaii.edu> wrote:
> senorb...@mail.utexas.edu wrote:
> : In article <6obfrh$ph8$4...@zingo.tninet.se>,
> : > >"longer" noun /longer/
> : > >"longer" adj /longre/
> : > >"sin" /sin/
> : > >"sing" /sing/
> : It _has_ arisen. Go ask any man on the street if he hears a difference
> : between [sIn] and [sIN], without any context that might give away the
> : meaning and making sure those vowels are the same, and I'm sure you'll get
> : some ...
>
> I certainly never said or implied that "sin" and "sing" might be
> phonemically the same. If you will just look above, you'll see
> two different phonemic forms for them, but neither with /N/.
I realize that, but the fact is that there is no phonetic segment [g] which
*might* (and I stress this highly) be said to allow the preceding nasal to be
merely a allophone of /n/. The fact is that, by any standard definition of
what a "phoneme" might be, the phone [N] is in this case a separate phoneme
/N/. To say otherwise is to ignore the data, and the way in which the
overwhelmingly large number of English speakers speak (and I only mean to be
forceful about this, not rude).
> : response that they are indeed different sounds. This is all you need to do
> : show that it's a phoneme: find what might be a minimal pair and ask a native
> : speaker.
>
> Even if it were a minimal pair, that doesn't prove the differing
> segments are different phonemes. In the /N/-less analysis, they
> derive from different phoneme sequences.
What makes something a phoneme is the fact that the people who use the
language natively *hear*the*difference*. Minimal pairs are one way of showing
that such a difference is made (though not so much in modern orthographies,
which are for the most part standardized). If you were to ask some
semiliterate child to write down the two words, I am quite positive that he
would ask to know how to write down the difference between the two sounds,
because it's part of the language. This fact just can't be avoided.
senorb...@mail.utexas.edu
"Peter T. Daniels" <gram...@worldnet.att.net> wrote:
> > There is also, in my dialect, a slightly different vowel before
> > all velar nasals. My /I/ in such circumstances is much closer to
> > /i/ in many ways. But I don't know how far that runs, in general
> > American dialects (or British ones, for that matter).
>
> I don't have that. Are you maybe hearing some sorts of diphthongs, as is
> much more frequent in Texas than in various other varieties?
It's certainly possible. If so, it would have to be something like /Iy/,
which for some reason seems strange to me. I still kinda think it's a vowel
somewhere in between [I] and [i], but I dunno. I'd have to check it
phonetically with some equipment at the phonetics lab on campus.
Thomas Anway
Someone proposed the following /N/-less hypothesis:
> : > >"longer" noun /longer/
> : > >"longer" adj /longre/ !!!!
> : > >"sin" /sin/
> : > >"sing" /sing/ !!
...
>
In the /N/-less analysis, they
> derive from different phoneme sequences.
Either propose a test to show that people really carry around a four
phoneme sequence for <sing> in their heads or concede it's a sophist's
exercise. (Carrying forth the theme that a hypothesis does not make a
scientific statement without evidence.) Simplicity favors three as a more
sensible hypothesis in the absence of evidence for a more complex model.
While you're at it, what's the evidence for /longre/ (sic)?
Occam's razor cuts through Chomsky-Hallean curlicues.
>
> --
> Greg Lee <l...@Hawaii.edu>
Name below is a pseudonym. I'm sure y'all can figure out who I am.
--
Thomas Anway - IMPORTANT remove caps from return mail address
to...@harlequin.co.uk
>Brian M. Scott <sc...@math.csuohio.edu> wrote:
>: On 15 Jul 1998 03:44:51 GMT, Greg Lee <l...@Hawaii.edu> wrote:
>
>: >When English speakers attempt to pronounce an [N] in a position
>: >where /ng/ could not occur, they can't do it, and substitute
>: >something else.
>
>: Statements like this bother me. This one is clearly false as stated;
>: is it actually intended as a statement of what *usually* happens?
>
>Yes, well. I'm an English speaker, and I can say [Nwin]. Many
>Vietnamese speakers also speak English ... It's hard to phrase it
>right. It's not what usually happens -- it *always* happens, but
>there are many qualifications that have to be understood. It has
>to be a "naive" English speaker, and he or she has to be attempting
>the pronunciation of a word, rather than imitating a sound.
>
Just what is a position in which /ng/ cannot occur? Or do you mean "does not"
rather than "could not"?
How about a word like "counting" - - wouldn't a "naive" English speaker assume
that /ng/ could not occur before t, so that this word is a direct
counterexample to your assertion?
Also, what do you mean by "attempting the pronunciation of a word rather than
imitating a sound"? Can a valid distinction be made between these two things,
in the case where the sound is in fact a word? If someone hears a word,
remembers it, and then uses it is one "imitating a sound" or "attempting the
pronunciation of a word"?
to...@harlequin.co.uk
>Thomas Anway <r...@NOSPMPLSE.monmouth.com> wrote:
>....
>: scientific statement without evidence.) Simplicity favors three as a more
>: sensible hypothesis in the absence of evidence for a more complex model.
>
>/N/?
>
Yes, of course it is.
An analysis without /N/ will have to invent all sorts of horrible workarounds
for the fact that N and /N/ appear to be recognised as distinguished in words
that are otherwise identical. What will the analysis do with pairs like
seen sing, been bing, keen king, bun bung, sun sung, ton tongue, con kong
(and so on: there are plenty more minimal pairs), Maybe one can argue that
/N/ arose originally as an allophone of N used before a palatal stop, and that
/N/ was not a phoneme in its own right at that time, but the stop in question
has long since been dropped from (most people's) pronunciation and a
description of tha language as it is spoken today and as its word distinctions
are recognised by hearers today cannot appeal to the (absent) stop. An
approach based on the existence of a silent allophone of g wich is used in the
ng context is going to run into big trouble on words like hunger, pengiun, and
so on, where the g is not silent, and the existence of pairs like longer (one
who longs) and longer (greater length) makes such an approach even more
complex. It is clearly going to be far simpler to accept /N/ as a phoneme than
to find othe ways of solving all these problems.
Greg Lee
: ... The fact is that, by any standard definition of
: what a "phoneme" might be, the phone [N] is in this case a separate phoneme
: /N/. ...
That is not a fact. It may, or may not be, a correct analysis.
: To say otherwise is to ignore the data, ...
What data is ignored by saying otherwise? Specifically.
: What makes something a phoneme is the fact that the people who use the
: language natively *hear*the*difference*. ...
But, specifically, what difference? A difference between words? Between
morphemes? Between phonemes? I think you're not being careful
enough to distinguish facts from theories.
...
: semiliterate child to write down the two words, I am quite positive that he
: would ask to know how to write down the difference between the two sounds,
Ok. And you'd tell him "n" versus "ng", right? Now if you could just
explain to me why you think this is an incorrect phonemic analysis ...
: because it's part of the language. This fact just can't be avoided.
--
Greg Lee <l...@Hawaii.edu>
Greg Lee
: English* (1968), you can get rid of some of the symbols (ex phonemes) in
: the underlying forms by making them conditioned variants of other
: symbols, even if they never surface, then you've achieved some precious
: economy. Hence the previously presented distinction between |longer| and
: |longre| (vertical brackets mark underlying forms in SPE).
Actually, the /longer/, /longre/ analysis was suggested to me by
David Stampe, though he might have been kidding.
There is another, better reason for doubting the existence of /N/.
When English speakers attempt to pronounce an [N] in a position
where /ng/ could not occur, they can't do it, and substitute
something else. Two stops can't occur in syllable onset in
English, and that's why (given the /ng/ analysis) "Nguyen" isn't
pronounced with initial [N] (well, syllabic [N]+[g] may be
possible). Phonetically, English is also consistent with a
process deleting syllable offset non-coronal voiced stops after
homorganic voiced stops (to get rid of [g] after [N]).
--
Greg Lee <l...@Hawaii.edu>
Keith C. Ivey
>> There is also, in my dialect, a slightly different vowel before
>> all velar nasals. My /I/ in such circumstances is much closer to
>> /i/ in many ways. But I don't know how far that runs, in general
>> American dialects (or British ones, for that matter).
>
>I don't have that. Are you maybe hearing some sorts of diphthongs, as is
>much more frequent in Texas than in various other varieties?
I don't know about other vowels, but I definitely hear this shift in
my speech for /I/ before /N/. Even in childhood, I felt that the
vowel in "sing" was closer to the one in "seen" than to the one in
"sin". But I don't think it's a diphthong, and it's certainly not a
Texas thing in my case.
Then there's Bob Cunningham in AUE, who's sure he has /eI/ before /N/
in "angle", "bank", and other words that most of us would transcribe
with /&/.
By the way, I still don't understand why eliminating /N/ as a phoneme
could be so important as to be worth introducing such complications as
transcribing "longer" (adj.) as /longre/. It reminds me of Ruud
Harmsen's sci.lang posts a while back trying to show that /T/ and /D/
were a really single phoneme in English.
Brian M. Scott
>When English speakers attempt to pronounce an [N] in a position
>where /ng/ could not occur, they can't do it, and substitute
>something else.
Statements like this bother me. This one is clearly false as stated;
is it actually intended as a statement of what *usually* happens?
Brian M. Scott
Peter T. Daniels
> There is another, better reason for doubting the existence of /N/.
> When English speakers attempt to pronounce an [N] in a position
> where /ng/ could not occur, they can't do it, and substitute
> something else. Two stops can't occur in syllable onset in
> English, and that's why (given the /ng/ analysis) "Nguyen" isn't
> pronounced with initial [N] (well, syllabic [N]+[g] may be
> possible). Phonetically, English is also consistent with a
> process deleting syllable offset non-coronal voiced stops after
> homorganic voiced stops (to get rid of [g] after [N]).
Phonological theorists keep trying to tell me "n" is a stop, and I
continue to not believe them.
Greg Lee
...
: scientific statement without evidence.) Simplicity favors three as a more
: sensible hypothesis in the absence of evidence for a more complex model.
An analysis of English with /N/ is simpler than an analysis without
/N/?
: While you're at it, what's the evidence for /longre/ (sic)?
I have no evidence -- it's ad hoc.
: Occam's razor cuts through Chomsky-Hallean curlicues.
So far as I know, Chomsky and Halle did not propose /longre/.
In the SPE framework, a more likely description would be
/long#er/ (noun) vs. /long+er/ (adj.). We should keep our
sins straight.
--
Greg Lee <l...@Hawaii.edu>
Greg Lee
: >When English speakers attempt to pronounce an [N] in a position
: >where /ng/ could not occur, they can't do it, and substitute
: >something else.
: Statements like this bother me. This one is clearly false as stated;
: is it actually intended as a statement of what *usually* happens?
Yes, well. I'm an English speaker, and I can say [Nwin]. Many
Vietnamese speakers also speak English ... It's hard to phrase it
right. It's not what usually happens -- it *always* happens, but
there are many qualifications that have to be understood. It has
to be a "naive" English speaker, and he or she has to be attempting
the pronunciation of a word, rather than imitating a sound.
--
Greg Lee <l...@Hawaii.edu>
Harlan Messinger
: There is another, better reason for doubting the existence of /N/.
: where /ng/ could not occur, they can't do it, and substitute
: English,
Ptooey!
Tore Lund
>
> /long#er/ (noun) vs. /long+er/ (adj.)
Since this is too advanced for me and my dictionaries, could someone
explain to me what the noun /long#er/ means?
--
Tore Lund <tl...@online.no>
Greg Lee
: I don't know about other vowels, but I definitely hear this shift in
: my speech for /I/ before /N/. Even in childhood, I felt that the
: vowel in "sing" was closer to the one in "seen" than to the one in
: "sin". But I don't think it's a diphthong, and it's certainly not a
: Texas thing in my case.
I have bar-I here -- lax high central unrounded vowel.
: Then there's Bob Cunningham in AUE, who's sure he has /eI/ before /N/
: in "angle", "bank", and other words that most of us would transcribe
: with /&/.
I have [@y] here -- low front vowel plus yod-glide.
: By the way, I still don't understand why eliminating /N/ as a phoneme
: could be so important as to be worth introducing such complications as
: transcribing "longer" (adj.) as /longre/. It reminds me of Ruud
Why is that a complication?
: Harmsen's sci.lang posts a while back trying to show that /T/ and /D/
: were a really single phoneme in English.
The impetus for such attempts is clear enough. We're dealing with
distinctions that are marginal (at best), and we know that languages
tend to minimize the sounds they use as phonemes. This is not a
methodological principle -- every human language has many, many
fewer phonemes than allophones. It's the way language works.
--
Greg Lee <l...@Hawaii.edu>
Brian M. Scott
>Greg Lee wrote:
>> /long#er/ (noun) vs. /long+er/ (adj.)
>Since this is too advanced for me and my dictionaries, could someone
>explain to me what the noun /long#er/ means?
One who longs (for), i.e., one who yearns (for). I don't think that
it's actually used, but the addition of agential <-er> to a verb is
productive, and the word is understandable.
Brian M. Scott
Tore Lund
Aaahh... y'all had me fooled here. To me, the noun "longer" seems
severely anomalous at the phonetic level. The "aw" sound refuses to
mesh with that [N@r.] ending. Just my gut-level feeling.
--
Tore Lund <tl...@online.no>
Peter T. Daniels
>
> In article <6oidm3$b...@news.Hawaii.Edu>, Greg Lee <l...@Hawaii.edu> wrote:
> >Thomas Anway <r...@NOSPMPLSE.monmouth.com> wrote:
> >....
> >: sensible hypothesis in the absence of evidence for a more complex model.
> >
> >An analysis of English with /N/ is simpler than an analysis without
> >/N/?
> >
> An analysis without /N/ will have to invent all sorts of horrible workarounds
> for the fact that N and /N/ appear to be recognised as distinguished in words
> that are otherwise identical. What will the analysis do with pairs like
> seen sing, been bing, keen king, bun bung, sun sung, ton tongue, con kong
> (and so on: there are plenty more minimal pairs),
The Great Midwestern Vowel Deficit once again rears its head ... seen
and sing, keen and king aren't minimal pairs for most English-speakers!
(Nor are been and bing for some speakers.)
Peter T. Daniels
We can also say (somewhat jocularly) "you can't be any wronger."
And I suppose there could also be a "belonger."
Peter T. Daniels
> Just what is a position in which /ng/ cannot occur? Or do you mean "does not"
> rather than "could not"?
> How about a word like "counting" - - wouldn't a "naive" English speaker assume
> that /ng/ could not occur before t, so that this word is a direct
> counterexample to your assertion?
Hunh?? Where is there an [N] before t here?
Every language has phonotactics, which say what sounds can appear where
(in a syllable, usually; sometimes in a word).
shr- is ok (shrimp), shl- isn't (it's only in foreign names!)
sl- is ok (slip), sr- isn't.
Italian can put all sorts of clusters at the beginning of a word that
English can't.
And N can't start a word in English; thus the Vietnamese name Nguyen is
usually not pronounced anything like the original.
Thomas Anway
> Thomas Anway <r...@NOSPMPLSE.monmouth.com> wrote:
> ...
> : scientific statement without evidence.) Simplicity favors three as a more
> : sensible hypothesis in the absence of evidence for a more complex model.
>
> An analysis of English with /N/ is simpler than an analysis without
> /N/?
A mental entry for <sing> with three symbols is simpler than four, yes,
which I'm sure you understood, but chose to ignore, along with my request
for evidence that doesn't exist that people really have four symbol
entries in their mental lexicons. It's also simpler to assume that the
underlying representation has the same number of elements as the
realization, eliminating an unnecessary processing step. Science is
supposed to be about reality, not unprovable and pointlessly complex
theories.
>
> : While you're at it, what's the evidence for /longre/ (sic)?
>
> I have no evidence -- it's ad hoc.
>
> : Occam's razor cuts through Chomsky-Hallean curlicues.
>
> So far as I know, Chomsky and Halle did not propose /longre/.
> In the SPE framework, a more likely description would be
> /long#er/ (noun) vs. /long+er/ (adj.). We should keep our
> sins straight.
>
> --
> Greg Lee <l...@Hawaii.edu>
--
Keith C. Ivey
>How about a word like "counting" - - wouldn't a "naive" English
>speaker assume that /ng/ could not occur before t, so that this
>word is a direct counterexample to your assertion?
Eh? How do you pronounce "counting"? I can't remember hearing anyone
say [N] before the /t/. It seems an unlikely pronunciation, since the
/t/ is presumably realized as something alveolar (or somewhere in that
neighborhood), whatever your dialect, and thus wouldn't be likely to
velarize the /n/.
Keith C. Ivey
>Tore Lund <tl...@online.no> wrote:
>>Since this is too advanced for me and my dictionaries, could someone
>>explain to me what the noun /long#er/ means?
>
>One who longs (for), i.e., one who yearns (for). I don't think that
>it's actually used, but the addition of agential <-er> to a verb is
>productive, and the word is understandable.
The noun "longer" is rare, but it is used. It's a hard thing to
search for, but I found these uses of the plural on the Web:
Nor are they so "hard-science" as the "longers for the Golden Age
of SF" would like to think or pretend.
(http://www.noveladvice.com/980201p1.htm)
And also the undertakers of all the religious rites who, by means
of the hope of rightful religion, render one certain as to the
way to the distant awful place, and tempt the longers for
righteousness into the religion, undertake all the religious
rites and ceremonial of the sacred beings for the sake of the
stipend of proper diligence.
(http://www.avesta.org/pahlavi/dd85.html)
Nor need one doubt that some sincere longers after God's fullness
were met by Him in grace, irrespective of what in the meetings
was not of Him, and there received a fullness of Holy Spirit not
known before.
(http://www.fardistant.demon.co.uk/tb/lang.htm)
Greg Lee
: Ptooey!
I'm a gnu.
I'm a gnu!
I'm the gnicest gnanimal in the gzoo.
I'm a gnu.
I'm a gnu!
And you really ought to know whwho's whwho!
--
Greg Lee <l...@Hawaii.edu>
Greg Lee
: Just what is a position in which /ng/ cannot occur? Or do you mean "does not"
In syllable onset, as I said. For example, at the beginning of a word.
: Also, what do you mean by "attempting the pronunciation of a word rather than
: imitating a sound"? Can a valid distinction be made between these two things,
A distinction can be made. I wouldn't presume to say whether it
was "valid". Articulate speech is circumscribed in ways that
imitative and expressive sounds are not.
: in the case where the sound is in fact a word? If someone hears a word,
: remembers it, and then uses it is one "imitating a sound" or "attempting the
: pronunciation of a word"?
Hard to say. If you would tend to report that he said "splurf",
probably a word, but if you reported that he went "splurf", then
just a sound.
--
Greg Lee <l...@Hawaii.edu>
Keith C. Ivey
>shr- is ok (shrimp), shl- isn't (it's only in foreign names!)
You don't consider "schlep" and "schlock" and "schlemiel" to be
English words? Which side of the pond are you on?
Keith C. Ivey
>We can also say (somewhat jocularly) "you can't be any wronger."
But doesn't that have a /g/, like "younger" and "longer"? I think it
does. I suppose someone who's wronged you could be a wronger
(/rON@r/), though. In fact, W9NCD lists "wronger" the noun (without a
/g/), and so does AHD3 (no pronunciation); neither has "wronger" the
adjective.
Brian M. Scott
wrote:
>"Peter T. Daniels" <gram...@worldnet.att.net> wrote:
>>We can also say (somewhat jocularly) "you can't be any wronger."
>But doesn't that have a /g/, like "younger" and "longer"?
Not in my speech.
> I think it
>does. I suppose someone who's wronged you could be a wronger
>(/rON@r/), though. In fact, W9NCD lists "wronger" the noun (without a
>/g/), and so does AHD3 (no pronunciation); neither has "wronger" the
>adjective.
The OED has only the noun; the pronunciation offered is essentially
/rON@r/.
Brian M. Scott
Brian M. Scott
>Brian M. Scott <sc...@math.csuohio.edu> wrote:
>: On 15 Jul 1998 03:44:51 GMT, Greg Lee <l...@Hawaii.edu> wrote:
>: >When English speakers attempt to pronounce an [N] in a position
>: >where /ng/ could not occur, they can't do it, and substitute
>: >something else.
>: Statements like this bother me. This one is clearly false as stated;
>: is it actually intended as a statement of what *usually* happens?
>Yes, well. I'm an English speaker, and I can say [Nwin]. Many
>Vietnamese speakers also speak English ... It's hard to phrase it
>right. It's not what usually happens -- it *always* happens, but
>there are many qualifications that have to be understood. It has
>to be a "naive" English speaker, and he or she has to be attempting
>the pronunciation of a word, rather than imitating a sound.
Yes, I took all that as read. But I'm still not convinced that it's
true. How naive is naive? The first time I saw the name, I was quite
young, and I knew nothing about Vietnamese, so I said what was there
on the page: [Nu jEn]. I was perfectly well aware that initial [N]
doesn't occur in English, but it offered no particular difficulty.
(It's certainly far easier than saying [mn@ 'mA nIk], which I did very
briefly before learning that the <m> was silent!)
Brian M. Scott
Tore Lund
>
> Tore Lund wrote:
> >
> > Aaahh... y'all had me fooled here. To me, the noun "longer" seems
> > severely anomalous at the phonetic level. The "aw" sound refuses to
> > mesh with that [N@r.] ending. Just my gut-level feeling.
>
>
Even if we conclude that [ON@r] actually exists, it does look like this
combination only occurs when some word in <ong> is supplemented by the
productive suffix <er>. And if that is so, would not this be an
indication that the English sound system is not really comfortable with
the sequence [ON@r]?
--
Tore Lund <tl...@online.no>
Greg Lee
: young, and I knew nothing about Vietnamese, so I said what was there
: on the page: [Nu jEn]. I was perfectly well aware that initial [N]
: doesn't occur in English, but it offered no particular difficulty.
: (It's certainly far easier than saying [mn@ 'mA nIk], which I did very
: briefly before learning that the <m> was silent!)
Then I'm wrong.
--
Greg Lee <l...@Hawaii.edu>
Greg Lee
: In article <6oidm3$b...@news.Hawaii.Edu>, Greg Lee <l...@Hawaii.edu> wrote:
: > An analysis of English with /N/ is simpler than an analysis without
: > /N/?
: A mental entry for <sing> with three symbols is simpler than four, yes,
: which I'm sure you understood, but chose to ignore, along with my request
There is an argument, which Peter alluded to, that the relevant
measure of simplicity is the inverse of the number of unpredictable feature
specifications in the entire lexicon plus those required to express
the phonological rules. Thus, you see, the question of which is
simpler does not really have an obvious answer. However, I don't
accept that simplicity has any relevance to which analysis is correct.
: for evidence that doesn't exist that people really have four symbol
: entries in their mental lexicons.
I know of no evidence that bears specifically on the number of
phnemes in "sing".
: It's also simpler to assume that the
: underlying representation has the same number of elements as the
: realization, eliminating an unnecessary processing step.
No, it isn't. Not, at least, in any sense of simplicity that could
possibly be relevant to this issue.
: Science is
: supposed to be about reality, not unprovable and pointlessly complex
: theories.
:
--
Greg Lee <l...@Hawaii.edu>
Peter T. Daniels
>
No; Michael Feldman says it every time a contestant on the "Whad'ya
Know" quiz gets two wrong answers ("You can't get any wronger--you can't
get any more wrong"). (Milwaukee native; public radio Saturdays 10-12
CST. Delayed broadcast, alas, in NYC 1-3 pm, so we can't call in and
participate.)
Peter T. Daniels
>
> You don't consider "schlep" and "schlock" and "schlemiel" to be
> English words? Which side of the pond are you on?
They are (a) highly marked as Yiddish borrowings--"Yinglish"--and hence
(b) belong to a (Firthian, i.e. other side of the pond!) distinct
phonological subsystem and (c) are not part of the vocabulary of the
vast majority of English-speakers.
And there's no reason to spell them with <c> as if they came
from/through German. Cf. the discussion of <shwa> vs. *<schwa> a while
ago.
Alwyn Thomas
> Even if we conclude that [ON@r] actually exists, it does look like this
> combination only occurs when some word in <ong> is supplemented by the
> productive suffix <er>. And if that is so, would not this be an
> indication that the English sound system is not really comfortable with
> the sequence [ON@r]?
Whose English sound system? There are plenty of people who say [lON@r]
(i.e. "more long") as a matter of course.
Alwyn
Keith C. Ivey
>Keith C. Ivey wrote:
>> "Peter T. Daniels" <gram...@worldnet.att.net> wrote:
>>
>> >shr- is ok (shrimp), shl- isn't (it's only in foreign names!)
>>
>> You don't consider "schlep" and "schlock" and "schlemiel" to be
>> English words? Which side of the pond are you on?
>
>They are (a) highly marked as Yiddish borrowings--"Yinglish"--and hence
>(b) belong to a (Firthian, i.e. other side of the pond!) distinct
>phonological subsystem and (c) are not part of the vocabulary of the
>vast majority of English-speakers.
It seems to me that Yiddish borrowings have affected the US English
phonological system to the extent that /Sl/ and /Sm/ and /Sn/ should
be counted as possible initial clusters. Certainly "schlemiel" isn't
part of most Americans' vocabularies, but I think "schlock" is pretty
widespread (in the US, at least) and that many people who use it don't
think of it as Yinglish. I have no evidence to support my belief,
however. ("Schlep" is somewhere between "schlemiel" and "schlock" in
Englishness, I'd say.)
>And there's no reason to spell them with <c> as if they came
>from/through German. Cf. the discussion of <shwa> vs. *<schwa> a while
>ago.
I thought about that when I was writing them, but I decided to go with
Merriam-Webster's (and American Heritage's) preferred spellings, which
are how I generally see them spelled.
Keith C. Ivey
>Keith C. Ivey wrote:
>> "Peter T. Daniels" <gram...@worldnet.att.net> wrote:
>> >We can also say (somewhat jocularly) "you can't be any wronger."
>>
>> But doesn't that have a /g/, like "younger" and "longer"?
>
>No; Michael Feldman says it every time a contestant on the "Whad'ya
>Know" quiz gets two wrong answers ("You can't get any wronger--you can't
>get any more wrong"). (Milwaukee native; public radio Saturdays 10-12
>CST. Delayed broadcast, alas, in NYC 1-3 pm, so we can't call in and
>participate.)
("Whad'ya Know" broadcast live from the Smithsonian on the Fourth.
That was the first time I'd heard it in quite a while.)
If "wronger" the adjective doesn't have /g/, that makes the /N/-less
analysis even more complicated, since there has to be some rule to
explain why /rongre/ is pronounced [rONR] while /longre/ is [lONgR].
I must be missing some other benefit that /N/-lessness produces.
Surely at some point one has to throw out the epicycles and decide
that things work better when the earth revolves around the sun.
Keith C. Ivey
>Even if we conclude that [ON@r] actually exists, it does look like this
>combination only occurs when some word in <ong> is supplemented by the
>productive suffix <er>. And if that is so, would not this be an
>indication that the English sound system is not really comfortable with
>the sequence [ON@r]?
No doubt there are many four-phoneme (or five-phoneme) sequences that
occur in only a few words. I don't think that means the English sound
system is uncomfortable with them. The combination /ON@r/ is no
stranger to this native English speaker than /IN@r/ or /&N@r/. People
don't use "longer" the noun much, but I don't think they have any
difficulty with it when they run across it, as in the sentences I
posted from the Web.