Re: [G] and [g] and PIE voiced plosives

--- In cybalist@yahoogroups.com, "tgpedersen" <tgpedersen@...> wrote:

>
> --- In cybalist@yahoogroups.com, "Andrew Jarrette" <anjarrette@> wrote:
> >
> > --- In cybalist@yahoogroups.com, "Andrew Jarrette" <anjarrette@>

wrote:

> > >
> > > --- In cybalist@yahoogroups.com, "tgpedersen" <tgpedersen@> wrote:
> > > > I was wondering if the different outcomes of Engl. -ough was
> > > > caused by them being different from the beginning, thus
> > > > -ough /-oUx/ > /-of/ and
> > > > -ough /-oUG/ > /-oU/
> > > >
> > > >
> > > > Torsten
> > > >
> > >
> > > That's what I would think too, but the evidence doesn't present a
> > > clear-cut pattern:
> > >
> > > <enough> [inVf] from OE <genog> [jeno:x] or [j@...:x] with final
> > > /-x/, inflected <genoge> [jeno:Ge] or [j@...:G@] etc. > <enow>
> > > [inaU] archaic
> > > plural of <enough>, also = <enough>
> > > <tough> [tVf] from OE <toh> [to:x] with final /-x/
> > > <rough> [rVf] from OE <ruh> [ru:x] with final /-x/
> > > <cough> [kAf],[kOf] from OE *<cohhian> with /xx/
> > > <trough> [trAf], [trOf] from OE *trog [trOx] with final /-x/
> > > <laugh> [læf] from OE <hlæhhan> with /xx/
> > >
> > > BUT
> > > <though> [DoU] from Scand. *To:x, *Tox, with final /-x/
> > > <dough> [doU] from OE <dag> [dA:x] with final /-x/
> > > <bough> [baU] from OE <boh> [bo:x] with final /-x/
> > > <slough> [slaU] from OE <sloh> [slo:x] with final /-x/
> > > <plough> [plaU] from late OE <ploh> [plo:x] with final /-x/
> >
> >
> > Oh the shame! I wrote <boh> instead of <bog> and <ploh> instead of
> > <plog>, basing them on alternative but etymologically wrong
> > spellings.
>
> Now, now, I do that all the time.
>
> > With this in mind perhaps there's a tendency for words ending in
> > *-g to develop diphthongs while those ending in *-h develop [-f];
> > *trog and *þoh/þo:h (Toh/To:h) are exceptions. Cybalist members
> > will have noticed these mistakes and corrected them before this
> > message gets posted.
>
> DEO says Da. dog, Sw. dock are loans from MLG doch, which around 1400
> replaced ODa. tho: < ON þó; ODa. tho: is preserved in the Jysk
> initiating particle / interjection 'to'. That leaves Du. 'toch'
> unexplained.
>
> I was wondering if these g-/w- and g-/j- pairs
> http://tech.groups.yahoo.com/group/cybalist/message/61985
> (in which the w- and j- part seems to be the etymologically correct
> one) could reflect an insecurity between 'hard' and 'soft'
> pronunciation of /g-/ ?
>
>

I doubt it. Note that the ME <go:l> "yule" you have listed actually
has yogh as the first character, which in ME represented [j], hence
this word always had [j] in English and PGmc, never [g] or [G].

The alternation between <jesan/gesan> and <jehan/gehan> is explained
by the 2nd and 3rd person sg. pres. indic. forms, where *jisiþ, *jihiþ
had [j] before [i]: by dissimilation [j] tends to be hardened in this
position, to [J] the palatal plosive, or to the palatal fricative
corresponding to the velar fricative [G]. In any case, these sounds
merge with the allophones of *g ([g] or [G]) which occur before *i(:)
in OHG, which are palatal, hence are confused with these sounds and
taken to be allophones of *g rather than of *j. Then the entire
conjugation of the verb is reinterpreted as having *g instead of *j.
Then derivative words such as *jistiz, *jestaz are also influenced by
the new reinterpretation of the verb, and are reinterpreted as
*gistiz, *gestaz (where *gistiz would sound the same as *jistiz due to
natural phonetic processes, while *gestaz would be an entirely new
formation with a consonant different from that of *jestaz (except in
some dialects and in OS where these two would sound the same also)).

The alternation between <g> and <w> in many of the words listed could
easily be coincidental, or just possibly could hint at a divergent
development of an original *gWH that I mentioned before, i.e. to [g]
before IE rounded vowels and to [w] before IE unrounded vowels. I
doubt that they're an indication of an alternation between [g] and
[G], because in many cases the alternation <g>~<v>/<w> occurs within
the same language, thus indicating that [w] was not an eventual
pronunciation of *g in this position.

Andrew