ELTE SEAS

Discussion Forum
It is currently Sun Nov 19, 2017 9:27 am

All times are UTC + 1 hour




Post new topic Reply to topic  [ 30 posts ]  Go to page Previous  1, 2
Author Message
 Post subject:
PostPosted: Sun Dec 16, 2007 8:20 pm 
Offline

Joined: Wed Nov 10, 2004 6:15 pm
Posts: 808
What exactly do you mean by ``stay in place''? Scheer does suppose that word-final empty nuclei do license, while word-internal ones do not (cf. French having tr/bl/etc. word finally, but not preconsonantally.) But in the Coda Mirror theory both word-final and coda consonants are unlicensed, hence likely to undergo lenition.

_________________
szp


Top
 Profile  
 
 Post subject: nem értem #1
PostPosted: Sun Dec 16, 2007 7:27 pm 
Offline

Joined: Fri Apr 27, 2007 9:37 pm
Posts: 17
Ségéral et Scheer (1999) on segmental health (p23)

If a coda consonant gets no licensing, how will it stay in place? Is there a difference between empty nucleus licensing, which just sticks the C into word final position and nucleus licensing, which sustains its segmental health?

Thank you for your answer.

_________________
Prince & Smolensky 2008 - Optimalizing America!


Top
 Profile  
 
 Post subject:
PostPosted: Wed Dec 05, 2007 9:01 pm 
Offline

Joined: Wed Nov 10, 2004 6:15 pm
Posts: 808
I have talked about each of the papers assigned. We touched upon Harris's licensing inheritance theory, Ségéral & Scheer's coda mirror, Lowenstamm's word-initial empty CV site, and his idea of branching onsets. So I do not see why you think it should be useless to read these papers. You may be required to compare analyses of a given phenomenon, for which you have to be aware of what these analyses are.

_________________
szp


Top
 Profile  
 
 Post subject: exams!
PostPosted: Wed Dec 05, 2007 2:59 pm 
Offline

Joined: Fri Apr 27, 2007 9:37 pm
Posts: 17
Question concerning what nature the exam we face will take up. Dear Sir, I mean, we did not follow the required reading as closely as we did in the GP class so maybe the exam would take a different shape? (Also, it seems a bit ludicrous to force us to fastidiously scrutinize papers you yourself don't agree with, I mean I know, know thy enemy but the ten times sixteen definitions of government probably already threw many to the fringes of insanity in the GP class why make matters worse?) Giving us a complex task to solve still seems much harder both for the teacher and the pupils so I cannot possibly have hopes for that.

_________________
Prince & Smolensky 2008 - Optimalizing America!


Top
 Profile  
 
 Post subject:
PostPosted: Tue Nov 27, 2007 7:56 am 
Offline

Joined: Fri Apr 27, 2007 9:37 pm
Posts: 17
partly in line with your answer meanwhile I came to the conclusion that CV could best be regarded a typological contraption - well, maybe I am going too far from observations based on just one case study.

_________________
Prince & Smolensky 2008 - Optimalizing America!


Top
 Profile  
 
 Post subject:
PostPosted: Tue Nov 27, 2007 12:37 am 
Offline

Joined: Wed Nov 10, 2004 6:15 pm
Posts: 808
So we can conclude that Broca's aphasiacs do not (always) distort syllable structure in the direction of less marked structures. If we want to base our markedness criteria on the production of these people, then we are likely to face serious difficulties in other areas, like syllable typology, as you also mention.

_________________
szp


Top
 Profile  
 
 Post subject: reality of CV
PostPosted: Mon Nov 26, 2007 5:24 pm 
Offline

Joined: Fri Apr 27, 2007 9:37 pm
Posts: 17
Turning back to the grim, dismal sombreness of the conundrums of phonological theory, I'd suppose KLV90 claim that CV is the least marked syllable type from which it follows, due to the nature of the CV/VC related theories (their minimalist complexity conditions) that every language has CV in some underlying sense or another.

What I mean by this is that a) observably: reduplication and reduction patterns should allow us to recognize the familiar CV pattern emerging from the mist of the language and b) connected to this: for me it seems that basically what CV and VC essentially claim is that languages, under fair and square circumstances have CVCVCV OR vCVCVCVc. When this is not the case we have to turn to various devices to explain the reasons, such as government and lincensing, etc.

With this in mind I sat down to inspect a script of an interview with an aphasic patient with a lesion in the Broca area (production affected mostly). The script consisted of sentences and the patient's attempts to repeat them. My predictions where that (v)CV(c) (the unmarked syllable) will emerge especially in the case of sentences the complexity of which are unmountable for the patient.

However my predictions weren't supported by the data: it seems analogies with previous syllable structure pieces and suchlike overwrote any markedness issue, such that syllable structure in certain cases got less marked but in other cases it got more marked (sometimes even with cross-linguistically avoided Cclusters). This is fairly problematic for any theory reclining upon syllable markedness, both OT-s and VC.

I mean we all take the urine out of minimalist syntax because it tries to explain aphasic behaviour with syntactic tree representations (the infamous Tree Pruning Hypothesis, cf. Friedmann 1994) but at least it tries.

_________________
Prince & Smolensky 2008 - Optimalizing America!


Top
 Profile  
 
 Post subject:
PostPosted: Sun Nov 18, 2007 6:52 pm 
Offline

Joined: Fri Nov 12, 2004 10:34 am
Posts: 97
Antigone, Ismene, as well as the two dead brothers are all siblings, therefore they form a constituent. Therefore we must assume a super-heavy syllable which contains two vowels and two consonants in its rhyme. The onset, then, must be the girls' uncle, Kreon.

[[O Kreon] [R [N Antigone, Ismene] [C Eteocies, Polynices]]]

Alternatively, we may assume that the four siblings are elements of a segment governed and thus lenited by Kreon, which is why it loses two of its elements.


Top
 Profile  
 
 Post subject:
PostPosted: Sun Nov 18, 2007 1:51 am 
Offline

Joined: Fri Apr 27, 2007 9:37 pm
Posts: 17
Though the factorial typology is rendered in a charming way, problems still stay on the scene. Je ne veux pas rendre Péter Rebrus triste but the Richness of the Base principle (Prince & Smolensky 1993) prohibits us referring directly to Zeus or anybody else from the Greek panthenon.

I'd suggest the constraint F[divinity_left_to_sonorant] (Siptár & Szentgyörgyi forthc.) instead. An alternative solution would be to apply the CVCVCVC phonological model to this (Recski & Szeredi forthc.): in CVCVCVC phonology, C-s and V-s ensuing each other consist a Moebius strip thus erasing all issues concerning word boundaries.


Top
 Profile  
 
 Post subject:
PostPosted: Sun Nov 18, 2007 12:05 am 
Offline

Joined: Fri Nov 12, 2004 12:09 pm
Posts: 8
Location: budapest
huncut felhocske wrote:
I still maintain that they cannot bury him since then Creon couldn't govern.

i'm still working on the problem, no publishable solution yet, but can promise groundbreaking revelations concerning ismene's debatable role and sophocles' hints for vc phonologists.

and one more idea, in brackets:
faithfulness constraint 1: obey divine laws and bury the dead or fear the arrows of zeus
faithfulness constraint 2: obey human laws and respect whoever comes to the throne

__________|___divine laws _|__human laws_____
:)_antigone_|_____________ |____*____________
__ ismene__|______*!______|_________________

why antigone is the moral heroine...

_________________
i played pool with john harris in a czech strip club.


Top
 Profile  
 
 Post subject:
PostPosted: Sat Nov 17, 2007 9:54 pm 
Offline

Joined: Fri Apr 27, 2007 9:37 pm
Posts: 17
I still maintain that they cannot bury him since then Creon couldn't govern.
Also, it's radical CV not strict CV, with respect to my previous rant.


Top
 Profile  
 
 Post subject:
PostPosted: Sat Nov 17, 2007 9:37 pm 
Offline

Joined: Fri Nov 12, 2004 12:09 pm
Posts: 8
Location: budapest
i have a little bit to say on the subject. my apologies in advance.
here it is:

Dramatis Personae:
Vantigone
Vismene
Cpolyneices
Ceteocles

Scene I
ViCpVaCe

Vantigone
Consider if thou wilt share the toil and the deed.

Vismene
In what venture? What can be thy meaning?

Vantigone
Wilt thou aid this hand to lift the dead?

Vismene

Thou wouldst bury him,-when 'tis forbidden to Thebes?

Vantigone
I will do my part,-and thine, if thou wilt not,-to a brother. False to him will I never be found.

(to be continued)

_________________
i played pool with john harris in a czech strip club.


Top
 Profile  
 
 Post subject: VCVCVCVCV
PostPosted: Fri Nov 16, 2007 10:23 pm 
Offline

Joined: Fri Apr 27, 2007 9:37 pm
Posts: 17
I was wondering whether anybody worked with van der Hulst's strict CV in a CV phonology frame. Just read the article on the former (1994 I guess) and it seems particularly plausible: all those C-s and V-s as far as the eye can see...


Top
 Profile  
 
 Post subject:
PostPosted: Sat Nov 03, 2007 10:09 pm 
Offline

Joined: Fri Apr 27, 2007 9:37 pm
Posts: 17
Thank you for your time & answers.

(I hope we will get another comment within a few years and a hot debate can evolve in a manner of decades. - in the meantime I continue working on my questions.)


Top
 Profile  
 
 Post subject: ANN-341.23 VC phonology
PostPosted: Sat Nov 03, 2007 9:09 pm 
Offline

Joined: Wed Nov 10, 2004 6:15 pm
Posts: 808
Rácz Péter wrote:
Dear Don Szigetvari,

I have a few questions on which I would like to know the stand of VC
phonology. I originally started to collect them so as to corner you next
class, as you expressed your desire last time that we asked questions.
However, they have now grown beyond the boundaries of regular seminar
pestering
You should nevertheless have tried in class: others might be interested.
Rácz Péter wrote:
therefore I'd be much grateful if you could spare some time
for a written answer. Some of these questions were suggested by a
renowned colleague of yours, however, I worked on them in detail so
considering them a verbatim copy&paste exercise would be a mistake.
Also, I've skipped some of your classes and only read the szabálytalan
fonológia chapter on VCVC so some of these questions might be already
redundant - for that I apologize.

I. What can we do about syllabic consonants?

I could, in a way following the generative phonologist tradition, come
up with utterly inconspicuous, godforsaken languages like Imdlawn
Tashlhiyt Berber which has syllabic allophones for every consonantal
segment maybe with a few exceptions and which scared Prince & Smolensky
so much they founded OT. Still there are numerous cases like the Slavic
syllable-forming /r/ that require analysis anyway and cannot be regarded
as exceptions to some universal patterning (not that I know anything
about Slavic languages). Syllabicity, as we know, cannot be encoded in
the melodic representation:

*V
|
'R'

- but then two consecutive V slots will have to be silenced as CCC
appears on the surface. Even if that would be possible (with siamese
twin burials or some other ethnographic marvel) that would collapse
languages with trinary C clusters (like, maybe(???) Georgian) and
syllabic consonants together into the same representation, and so on.
Syllabicity of English sonorants is clearly represented as follows:
Code:
V C
\|
`R'

That is the əC sequence changes only in that the C spreads out over the empty (hence schwa) V part of the skeleton. A syllabic consonant functions like an unstressed vowel, i.e., a live vowel capable of governing: e.g., you can have syncope before it (national [næʃ.nḷ]), but also like a consonant: e.g., the same restrictions apply on syllabic consonant formation after a syllabic consonant as to a nonsyllabic consonant (indolence [ɪn.dl.əns], but *[ɪn.dl.ns] -- since [n] is less sonorous than [l]). This follows from their double linkage.

Apparently, there are languages (Danish is one) in which (some) syllabic consonants have to be represented as follows:
Code:
C V
|/
`R'

I have not examined the options in much detail yet.
Rácz Péter wrote:
II. Maybe this goes partly to the above question (my judgements on
consonant syllabicity are slightly below average): what could we do with
syncope in /firmament/ (In a rothic accent)?

1 2 3 4 5 6
VCVCVCVCVCVC
| | | | | | | |
f i r m m e n t

V2 takes care of V1, V6 is buried by the RT (well, here NT) cluster, and
we could either switch off V3 or V4 but not both. (Vegyük észre that
both /firm/ & /ment/ are well-formed syllables in English, not that this
would amount up to anything.) This might be just a little exception to
the general rule, like /ptato/ but it can be a symptom as well.
  1. Wells says syncope is not possible in this word.
  2. If it were
    1. the v between r and m would be buried, the one between the two m's would be governed, just like in any postcluster syncope (cf. comp<a>ny)
    2. the coda r is linked to a V in most English dialects (also see next question)
Rácz Péter wrote:
III. Furthermore, what to do with rothic/nonrothic? Do we argue that in
BrE /bar/ the vowel is long and two V slots licence each other (I mean
one licenses the other) which would mean that the rothic representation
is notably different:

VCVCVC vs VCVC
| ? | | |
b a b a r

--- this pretty much amounts up to the Classical GP analysis using
allophones to explain even inflectional paradigms (as in Kaye's
Derivations and Interfaces*)
or could it be proposed that the second slot of the long /a/ (same case
with /o/ in RP and as far as I know with all 'broken' vowels in spoken
British English) hangs from a C slot?

*Which is no sin with clearly irregular patterns like /keep-kept/ but it
is more than a sin; an error with regular paradigms in languages which
have inflectional paradigms at all (i.e. not English nor Chinese)
I think the following are the representations of nonrhotic and rhotic bar:
Code:
v C V c V c
  | |  /
  b A

v C V c V c
  | |  /
  b AR
where R denotes rhoticity, an optional property of nonhigh vowels.

There is plenty of reason to believe that in most English rhotic dialects the vowel and the following coda R are in a branching nucleus (cf. Harris 1994, ch.5).
Rácz Péter wrote:
IV. If we take care of #CC with some sort of consonantal licensing
relation (this is only hinted at in the szabálytalan book) that would
mean that #TR and VTRV are granted birth by different forces (the former
by licensing the latter by government). Are there any observations
confirming these predictions (like that the two groups act differently?
Putting aside the border signal-ness of #TR for a moment)? (#RT and TR#
clusters are - so it seems - banned by VC phon. with which I am kind of
comfortable though surely there is some language somewhere where they
grow in abundance.)
Honest answer: no idea. The best solution seems to be to treat branching onsets as monosegmental, but this has drawbacks. More on this on 22 Nov. according to the schedule.
Rácz Péter wrote:
Tobias Scheer, whose name I managed to avoid mentioning for three whole
questions, handles the problem with infrasegmental government: his
theory is then able to predict which #CC clusters are possible and which
aren't (he also tricks a bit with the presence/absence of word-initial
empty CV) - with some additional juggling this could be made
language-specific to fit the Icelandic<<Georgian "sonority" scale.
I criticize what he says on pp. 114ff of my diss which is available on-line. The most serious argument against his original (1996) solution is that it violates locality.
Rácz Péter wrote:
V. Scheer. He highlights a problem of Classical GP in his bible that GP
cannot handle languages that show word-final coda effects: so that the
wf C acts as if it was in the same syllable as the preceding V therefore
triggering processes which do happen in 'coda/pre-coda' environments:
his examples are Brazilian Portugese /l/-vocalisation and Turkish closed
syllable shortening. GP considers word-final C-s as onsets followed by
empty nuclei and even masterminds like John Harris (he in his textbook)
deny the existence of languages showing the opposite pattern (as above)
-n.b. he doesn't deny the existence of the languages but that they do that.
The GP analysis, as Scheer shows, has more wounds but GP is not to be
discussed here. He, of course, suggests a parameter on the licensing
ability of final empty nuclei which VC, conspicuously lacking the latter
contraptions, cannot do (putting aside at present the question whether
it is welcome to dilute a theory with principles switchable on to off).
I don't see why standard GP could not parameterize the moraicness of word-final codas just like Scheer. Anybody who has word-final empty nuclei can do this. Those who don't can't.
Rácz Péter wrote:
These are the few things I would like to ask. At the moment. I hope most
of my questions made sense in one way or another. (I am in fact a
treasurehouse of questions. However I already had time to note that the
more general ones ("why sonority?", "why analogy?", "why sounds?" (why
triple embedding?), etc.) receive woebegone shrugs from veteran
phonologists.)

Your humble servant,
rp
Thank you for your questions. I hope they will incite further discussion which others in the group will not be able to resist to join.

_________________
szp


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 30 posts ]  Go to page Previous  1, 2

All times are UTC + 1 hour


Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group