Re: Some accentological thoughts...

--- In cybalist@yahoogroups.com, Miguel Carrasquer Vidal <miguelc@...>
wrote:

>
> On Tue, 27 Feb 2007 12:39:49 -0000, "tgpedersen"
> <tgpedersen@...> wrote:
>
> >--- In cybalist@yahoogroups.com, Miguel Carrasquer Vidal <miguelc@>
> >wrote:
> >>
> >> On Mon, 26 Feb 2007 22:53:43 -0000, "tgpedersen"
> >> <tgpedersen@> wrote:
> >>
> >> >> The solution is to start from a mobile paradigm
> >> >> *vIdová, acc. vÍdovoN. Then, weak yers lose their
> >> >> stressability: *vIdová, *vIdóvoN.
> >> >
> >> >In terms I can understand:
> >> >*vidová, *vídovoN -> (jerification)
> >> >*vIdová, *vídovoN -> (regularization)
> >>
> >> I wouldn't put it like that. The change i > I was
> >> unconditional and independent of stress.
> >>
> >> >*vIdová, *vIdóvoN
> >
> >The *result* of the rule, yes. But was it applied everywhere
> >simultaneously? We don't know. This mechanism ultimately delivers
> >the same result. There's something that doesn't feel right about
> >the idea of a vowel that gets so short that it can't carry the
> >stress anymore. Why should that happen to a stressed vowel?
> >Jerification of unstressed vowels makes phonological sense.
> >Jerification of stressed vowels doesn't, so we better find a
> >different mechanism which achieves the same result.
>
> The mechanism to look for is the one that explains the
> attested facts. And the facts are that Proto-Slavic /i/ and
> /u/ were "reduced" (to something like /'&/, /&/) everywhere,
> independently of the stress. Then, following Havlík's law,
> which is also independent of the position of the stress, the
> weak yers were further reduced to zero. The mechanism
> you're proposing would have produced cases of unreduced /i/
> and /u/ under the stress (e.g. in fixed-stress paradigms),
> which simply do not occur.

But the standard one you propose would lead to stressed schwa's in
those fixed-stress paradigms, and they don't occur either. I didn't
propose to do away with Havlík.

> Phonologically, there is nothing nonsensical about stressed schwa's.
> They exist in many languages, such as Catalan, Romanian, English,
> Slovene, Bulgarian, Sanskrit, etc.

Irrelevant. I never said anything about stressed schwa's.

> >The reconstructions of PIE stress in whichever nominal paradigms
> >I've seen had the same stress in Nsg and Asg. What is the reason
> >for the stress alternation eg golová, gólovu?
>
> In the vowel stems, mobility is secondary in Balto-Slavic.
> The mobility comes from the C-stems.
>
> The proterodynamic paradigm:
>
> PIE
> N *h2ák^mo:n
> A *h2ák^monm.
>
> Hysterodynamic:
>
> N *dhugHté:r
> A *dhugHtérm.
>
> In Balto-Slavic, two things happened that caused nom. and
> acc. stress to diverge. In the first place, a posttonic long
> vowel (lengthened grade, not two contracted vowels or a
> vowel + laryngeal) attracted the stress. Secondly, syllabic
> resonants developed into diphthongs R. > iR (uR).
>
> This way, we get:
>
> N *h2ak^mó:n *dhugHté:r
> A *h2ák^menim *dhugHtérim

Erh, where did that Acc -e- come from?

> Polarisation of the stress to either initial or final
> position gives Proto-Balto-Slavic:
>
> N *h2ak^mó~ *dhugHté~
> A *h2ák^menim *dhúgHterim
>
> = Lith. akmuõ, àkmeniN; dukte~, dùkteriN
>
> This was the kind of mobility that was transferred to the
> oxytone vowel-stems:
>
> N *dhroughós
> A *dhroughóm > *dhróughom

N drugó-, A drúgU ??

> N *ungnís
> A *ungním > *úngnim
>
> N *suHnús
> A *suHnúm > *súHnum

I suppose Havlík cleans up here?

> N *galHwáh2
> A *galHwáh2m > *gálHwah2m


> but of course not the neuters:
>
> NA *pteróm
>

The i- and u-stems are obviously not much evidence, nor the o-stems.
Wouldn't it be easier to assume *-áx was stressed?


Torsten