"Anne Marie"
anal...@hotmail.com
"Ammarie" (at least to my ears) .
Is this a case of lenition working on two consonants that are too
close together and fusing them into one?
anal...@hotmail.com
wrote:
> assimilation of the palatal nasal /n/ to the bilabial nasal /m/.
>
Thanks.
But I have also heard people who say the 'n' distinctly. Is the
assimilation of the "n" a result of the speakers seeking reduced
articulatory effort?
I have also heard pronunciations of "atlantic" which sound practically
(to me) as "allantic". Is a similar phenomenon at work?
In general, does sound change work on consonant clusters to either
insert a vowel between them or double them up?
> --
> Posted via a free Usenet account fromhttp://www.teranews.com
Nathan Sanders
<833af08f-94a9-4098...@i7g2000prf.googlegroups.com>,
anal...@hotmail.com wrote:
> On Apr 1, 7:02 pm, Christopher Culver <crcul...@christopherculver.com>
> wrote:
> > analys...@hotmail.com writes:
> > > I have heard many people say it so that it can't be told apart from
> > > "Ammarie" (at least to my ears) . Is this a case of lenition
> > > working on two consonants that are too close together and fusing
> > > them into one?
> >
> > It's a case of syncope where the unstressed second /a/ is lost, then
> > assimilation of the palatal nasal /n/ to the bilabial nasal /m/.
>
> Thanks.
>
> But I have also heard people who say the 'n' distinctly. Is the
> assimilation of the "n" a result of the speakers seeking reduced
> articulatory effort?
Speakers don't "seek" sound change (that implies conscious choice in
the matter). Sound change just happens, and speakers are along for
the ride.
In this case, nasals are particularly prone to assimilating the place
of articulation of a following consonant because the acoustic cues for
place of articulation are rather weak in nasals. Combine that with
the natural physical coarticulation that happens between adjacent
sounds, and the first nasal, even if truly pronounced as [n], will
sound a lot like [m] (especially in fast speech).
This is such a strong effect that numerous languages have developed
nasal place assimilation independently.
> I have also heard pronunciations of "atlantic" which sound practically
> (to me) as "allantic". Is a similar phenomenon at work?
In some dialects of English, /t/ in a coda is pronounced as [?], which
you may be interpreting as contributing to the length of the following
consonant.
> In general, does sound change work on consonant clusters to either
> insert a vowel between them or double them up?
Lots of things can happen. You can get epenthesis of a vowel,
deletion of the first consonant, deletion of the second consonant,
assimilation (which could be either partial or total and either
regressive or progressive), dissimilation, metathesis, or no change at
all.
And it will vary by language, by which consonants are involved, and by
what context they are in. A given language may have multiple
different ways of repairing illicit sequences of consonants. English,
for example, has at least six processes just for handling word-final
consonant sequences:
* epenthesis: classes, dishes, rented, prism (or perhaps this is
syllabifying the [m])
* delete C1: sign, phlegm (cf. signature, phlegmatic)
* delete C2: damn, hymn, bomb (cf. damnation, hymnal, bombadier)
* partial regressive assimilation: tenth, wealth (dental assimilation)
* partial progressive assimilation: dog[z] vs. dock[s], locked with
[t] vs. logged with [d]
* no change: false, six, forth
I can't think of any cases of dissimilation of features, though the
epenthesis for the plural and past tense endings are dissimilatory in
nature (it only happens when the two consonants are too similar).
There are also some cases of metathesis (dialectal "aks" for "ask",
modern English "wasp" < "wćps"), but these aren't productive in modern
standard English. In fact, I think the "aks" metathesis isn't
productive even in dialects that have "aks". It's a sporadic change,
with the productive change in those dialects for word-final /sk/
sequences being deletion of the /k/ (as in "desk" [dEs]).
Nathan
--
Nathan Sanders
Linguistics Program
Williams College
http://wso.williams.edu/~nsanders/
Christian Weisgerber
> > I have also heard pronunciations of "atlantic" which sound practically
> > (to me) as "allantic". Is a similar phenomenon at work?
>
> In some dialects of English, /t/ in a coda is pronounced as [?], which
> you may be interpreting as contributing to the length of the following
> consonant.
Seems farfetched. How about a lateral release of /t/ before /l/?
A while ago I came across this:
http://learnit.gocomp.com/pron.html
"[t] is the most complicated sound: in Standard American English it has
six variations. [...]"
I'm not sure how accurate this is, but it's interesting.
--
Christian "naddy" Weisgerber na...@mips.inka.de
Nathan Sanders
na...@mips.inka.de (Christian Weisgerber) wrote:
> Nathan Sanders <nsan...@williams.edu> wrote:
>
> > > I have also heard pronunciations of "atlantic" which sound practically
> > > (to me) as "allantic". Is a similar phenomenon at work?
> >
> > In some dialects of English, /t/ in a coda is pronounced as [?], which
> > you may be interpreting as contributing to the length of the following
> > consonant.
>
> Seems farfetched.
Which part? Coda /t/ is most certainly pronounced as [?] is some
dialects (my own idiolect is one), and a sequence like [?l] could
easily be mistaken by a layman as a lengthened l (which is what I
assume analyst meant to represent with the <ll>; if he simply meant
<ll> to mean a singleton [l], with no noticeable [t], then that too
would be a reasonable interpretation of a [?l] sequence], since
English doesn't actually have a geminate [ll]).
> How about a lateral release of /t/ before /l/?
That doesn't typically happen in this position (it happens when the
/t/ is in the onset of an unstressed syllable that has /l/ as its
nucleus, as in "battle"). In "Atlantic", the /t/ is in a coda, and
the /l/ is in the onset of a stressed syllable.
> A while ago I came across this:
> http://learnit.gocomp.com/pron.html
> "[t] is the most complicated sound: in Standard American English it has
> six variations. [...]"
> I'm not sure how accurate this is, but it's interesting.
Their terminology is sometimes wrong (they've misused, at minimum,
"plosive" and "implosive") and they got some of the details wrong
(e.g., /t/ is aspirated not only at the beginning of stressed
syllables, but also at the beginning of words, even if the first
syllable is unstressed: "terrific"). A quick glance at the rest of
the webpage reveals a similar lack in accuracy sprinkled throughout.
As for their list of allophones of /t/, they missed at least five:
* the dentalized variant before interdental fricatives (as in
"outthink" and "...at the...")
* the retroflex flap that occurs between /r/ and an unstressed vowel
(as in "party" and "parted"; this is distinct from the alveolar tap
when no /r/ precedes the /t/, as in "potty" and "potted")
* the (partially) rounded variant that occurs before /w/ (as in
"twenty" and "twin")
* the retracted/retroflex (and often affricated) variant before /r/
(as in "train" and "attract")
* the unpronounced variant (as in the many dialects in which
"scripts", "acts", "banter", and/or "mantle" rhyme with "Scripps",
"axe", "banner", and/or "panel", respectively)
John Atkinson
[...]
>
> There are also some cases of metathesis (dialectal "aks" for "ask",
> modern English "wasp" < "wćps"), but these aren't productive in modern
> standard English. In fact, I think the "aks" metathesis isn't
> productive even in dialects that have "aks". It's a sporadic change,
> with the productive change in those dialects for word-final /sk/
> sequences being deletion of the /k/ (as in "desk" [dEs]).
In the particular case of "ask", isn't the metathesis the other way
around? Old English had both <acsian> and <ascian>, and, according to
the Oxford, "Till c1600 _ax_ was the literary form". Are there any
dialects where it is well-established that "aks" isn't just the older
form and that in those dialects the metathesis _failed_ to take place --
as is apparently the case for "wops" in those same dialects?
Admittedly, I'm thinking in particular of English dialects, while I
assume you're thinking of US rural and Southern dialects -- but I'm by
no means sure that the situation isn't the same there -- as you say, in
many of those US dialects, the standard change is to drop a final stop,
so it makes sense that the "k" in "ask" was never final in those
dialects that have "aks".
John.
Brian M. Scott
<john...@bigpond.com> wrote in
<news:uUYIj.5422$n8....@news-server.bigpond.net.au> in
sci.lang,alt.lang:
> "Nathan Sanders" <nsan...@williams.edu> wrote
> [...]
>> There are also some cases of metathesis (dialectal "aks" for "ask",
>> modern English "wasp" < "wæps"), but these aren't productive in modern
>> standard English. In fact, I think the "aks" metathesis isn't
>> productive even in dialects that have "aks". It's a sporadic change,
>> with the productive change in those dialects for word-final /sk/
>> sequences being deletion of the /k/ (as in "desk" [dEs]).
> In the particular case of "ask", isn't the metathesis the other way
> around? Old English had both <acsian> and <ascian>, and, according to
> the Oxford, "Till c1600 _ax_ was the literary form". Are there any
> dialects where it is well-established that "aks" isn't just the older
> form and that in those dialects the metathesis _failed_ to take place --
> as is apparently the case for "wops" in those same dialects?
OE <acsian> was already the result of metathesis: <áscian>
is the original form, from PGmc. *aisk-. My reading of the
OED is that in general the 'aks' type continues this ancient
OE metathesis, while the standard type continues a northern
survival of unmetathesized <ascian> (with shortening of the
initial vowel). The non-northern reflexes of <ascian> were
<ash> and <esh>, which are now lost completely.
This isn't to deny the possibility of independent later
metatheses as well, of course. In particular, it wouldn't
surprise me to learn that there are relatively recent
instances of /sk/ > /ks/.
[...]
Brian
anal...@hotmail.com
> In article <ft0p0v$1ak...@kemoauc.mips.inka.de>,
> na...@mips.inka.de (Christian Weisgerber) wrote:
>
> > Nathan Sanders <nsand...@williams.edu> wrote:
>
> > > > I have also heard pronunciations of "atlantic" which sound practically
> > > > (to me) as "allantic". Is a similar phenomenon at work?
>
> > > In some dialects of English, /t/ in a coda is pronounced as [?], which
> > > you may be interpreting as contributing to the length of the following
> > > consonant.
>
> > Seems farfetched.
>
> Which part? Coda /t/ is most certainly pronounced as [?] is some
> dialects (my own idiolect is one), and a sequence like [?l] could
> easily be mistaken by a layman as a lengthened l (which is what I
> assume analyst meant to represent with the <ll>; if he simply meant
> <ll> to mean a singleton [l], with no noticeable [t], then that too
> would be a reasonable interpretation of a [?l] sequence], since
> English doesn't actually have a geminate [ll]).
Thanks for the amazingly detailed analysis.
If you go to the Merriam Webster dictionary site, in their suggested
pronunciation the "t" can be clearly heard.
\ət-ˈlan-tik, at-\
I think some kind of "reduced articulatory effort" (I know that you
disagree) is at work when you hear people fail to articulate the "t".
But if "Atlantic" came from "Atalanta" an unstable sound change seems
to have taken place by dropping the vowel between t and l which then
brought two consonants together which in its turn seems to be causing
a lenition towards not articulating the "t".
Peter T. Daniels
> But if "Atlantic" came from "Atalanta"
But it didn't.
> an unstable sound change seems
> to have taken place by dropping the vowel between t and l which then
> brought two consonants together which in its turn seems to be causing
> a lenition towards not articulating the "t".
Even if it did, the speaker would have no way of knowing that this -
tl- was any different from any other -tl-.
Harlan Messinger
Yet, out of all the possibilities, including "bask", "cask", "task",
"desk", "risk", and "tusk", the only one I know of that gets altered is
"asterisk", but that becomes "asterik", not "*asteriks", and I assume
that this syncope is due to the influence of the preceding [st].
Nathan Sanders
<e6690529-bf67-4ae8...@e10g2000prf.googlegroups.com>,
anal...@hotmail.com wrote:
> On Apr 2, 9:15 pm, Nathan Sanders <nsand...@williams.edu> wrote:
> > In article <ft0p0v$1ak...@kemoauc.mips.inka.de>,
> > na...@mips.inka.de (Christian Weisgerber) wrote:
> >
> > > Nathan Sanders <nsand...@williams.edu> wrote:
> >
> > > > > I have also heard pronunciations of "atlantic" which sound practically
> > > > > (to me) as "allantic". Is a similar phenomenon at work?
> >
> > > > In some dialects of English, /t/ in a coda is pronounced as [?], which
> > > > you may be interpreting as contributing to the length of the following
> > > > consonant.
> >
> > > Seems farfetched.
> >
> > Which part? Coda /t/ is most certainly pronounced as [?] is some
> > dialects (my own idiolect is one), and a sequence like [?l] could
> > easily be mistaken by a layman as a lengthened l (which is what I
> > assume analyst meant to represent with the <ll>; if he simply meant
> > <ll> to mean a singleton [l], with no noticeable [t], then that too
> > would be a reasonable interpretation of a [?l] sequence], since
> > English doesn't actually have a geminate [ll]).
>
> Thanks for the amazingly detailed analysis.
>
> If you go to the Merriam Webster dictionary site, in their suggested
> pronunciation the "t" can be clearly heard.
>
> \?t-?lan-tik, at-\
Note the phrase "some dialects" in my statement above. Coda /t/ in
standard American English in simply just unreleased (represented in
the IPA by the top-right corner diacritic, by [_}] in X-SAMPA, and by
[<o>] in Kirshenbaum, I think, if I'm reading his specification
correctly).
In many dialects, the unreleased /t/ is produced with a glottal
closure as well, and in some of those dialects, the alveolar
articulation is also lost, leaving only a glottal stop.
> I think some kind of "reduced articulatory effort" (I know that you
> disagree) is at work when you hear people fail to articulate the "t".
It's a reduction of the alveolar articulation, yes. But there is also
the added glottal closure to factor in. It's not just a
straightforward case of pure articulatory ease, because a new,
completely different (and thus, possibly incomparable) articulation is
being used.
(I don't know of any major dialects of English in which coda /t/ is
completely deleted, with no glottal closure at all.)
Paul J Kriha
news:65ka0fF...@mid.individual.net...
Actually, Harlan, I have heard people say "asteriks" around here.
Most of the people say "asterisk" but "asteriks" is heard now
and then in the local English computerdownunderese.
I don't remember though, if I also heard it in Britain.
pjk
Brian M. Scott
<hmessinger...@comcast.net> wrote in
<news:65ka0fF...@mid.individual.net> in
sci.lang,alt.lang:
> Brian M. Scott wrote:
[...]
>> OE <acsian> was already the result of metathesis: <áscian>
>> is the original form, from PGmc. *aisk-. My reading of the
>> OED is that in general the 'aks' type continues this ancient
>> OE metathesis, while the standard type continues a northern
>> survival of unmetathesized <ascian> (with shortening of the
>> initial vowel). The non-northern reflexes of <ascian> were
>> <ash> and <esh>, which are now lost completely.
>> This isn't to deny the possibility of independent later
>> metatheses as well, of course. In particular, it wouldn't
>> surprise me to learn that there are relatively recent
>> instances of /sk/ > /ks/.
> Yet, out of all the possibilities, including "bask", "cask", "task",
> "desk", "risk", and "tusk", the only one I know of that gets altered is
> "asterisk", but that becomes "asterik", not "*asteriks", and I assume
> that this syncope is due to the influence of the preceding [st].
I've heard 'asteriks' as well as 'asterik'. I believe that
[dEks] for 'desk' is not uncommon in the speech of young
children.
Brian
benl...@ihug.co.nz
wrote:
> "Harlan Messinger" <hmessinger.removet...@comcast.net> wrote in message
>
> news:65ka0fF...@mid.individual.net...
>
>
>
> > Brian M. Scott wrote:
> > > On Thu, 03 Apr 2008 04:18:34 GMT, John Atkinson
> > > <news:uUYIj.5422$n8....@news-server.bigpond.net.au> in
> > > sci.lang,alt.lang:
>
> > >> "Nathan Sanders" <nsand...@williams.edu> wrote
Is it too much to suggest the influence of the cartoon character?
Ross Clark
Trond Engen
> Is it too much to suggest the influence of the cartoon character?
I'll vote for that. 'Asterisk' is hardly a word in Norwegian outside the
field of philology -- I'm used to hearing 'stjerne' "star" for the
character -- but I've heard people speak of a standing stone as an
'obeliks'.
--
Trond Engen
- presenting his Idéfix
Paul J Kriha
news:e5c01d66-df5f-433a...@c19g2000prf.googlegroups.com...
:-)
When I was writing my earlier note, I did considered the influence
of his name.
I have been, myself, quite keen on Asterix books since the early
seventies. Over the years whenever I heard somebody say
"asteriks", which around the computers was quite often, if I had
the opportunity to do so I asked them about the Asterix books
and cartoons. As far as I remember, only a minority of the
asteriksites knew anything of the name of Asterix.
pjk
P.S. I don't recall hearing anybody say "obeliks", though.
Harlan Messinger
Whereas in the States that influence wouldn't arise because for the most
part Asterix's adventures are unknown, as are Tin-Tin's.
António Marques
Some of my co-workers (portuguese) have fun saying 'Asterix' /aStE'riks/
for 'Asterisk' ['aStIriSk] (www.asterisk.org). No one ever though of
doing the same with the native 'asterisco' [@StI'riSku] <*>, probably
because of the different vowels.
António Marques
I was shocked to know that the Superduck
(http://en.wikipedia.org/wiki/Paperinik) is all but unknown in the US
(and moreover, what little is known is through modern versions missing
the fun of the classic character).
Marc
> In some dialects of English, /t/ in a coda is pronounced as [?], which
> you may be interpreting as contributing to the length of the following
> consonant.
I don't think Christopher meant gemination, just the absence of any
indication that there used to be a 't' there, including a glottal
stop. "Alanic" for "Atlantic" seems very plausible in rapid speech.
Try the following on an unsuspecting bystander: "I went to Alanic City
last year." I doubt they'd bat an eyelash. "And?" is what you'd get in
response, most likely.
Marc
Bart Mathias
> [...]
>
> Try the following on an unsuspecting bystander: "I went to Alanic City
> last year." I doubt they'd bat an eyelash. "And?" is what you'd get in
> response, most likely.
Tsway Eng'lish works. Thirty-two years ago I used to amuse myself
trying to find someone who would question my reference to "speared a
senny six." No one ever.
Bart Mathias