--- In
cybalist@yahoogroups.com, "Patrick Ryan" <proto-
language@...> wrote:
>
> ----- Original Message -----
> From: elmeras2000<mailto:jer@...>
> To: cybalist@yahoogroups.com<mailto:cybalist@yahoogroups.com>
> Sent: Friday, May 20, 2005 6:52 PM
> Subject: Re: [tied] IE *de:(y)- 'bind'.
>
>
>
> <snip>
>
> > ***
> > Patrick wrote:
> >
> > And you do not think that has any bearing on *-to as a
future-
> prospective formant?
> > ***
>
> Not as I am informed at the moment, no.
>
> ***
> Patrick writes:
>
> In PIE, -*t(o) as an agentive formant of nouns is _exactly_ what
we should expect if -*to has future-prospective force in verbs.
By what principle? I see no rhyme or reason in this.
>
> I therefore consider the comparison I made among *mór-to-s,
Sumerian -ed (for *et), and Egyptian sDm.t.f, sDm.tj.fj, and
infinitives in .t validated.
> ***
I'm sure you do. I'm afraid I can't use your conviction for anything
interesting.
>
> >
> > JER:
> > > If the remarkable Indic forms with iraj- are related it
must
> be
> > a
> > > laryngeal.
> > >
> > > ***
> > > Patrick wrote:
> > >
> > > *H3o- as a prefix also contains a 'laryngeal'.
> >
> > You can't derive Indic ir- from *H3or-.
> >
> > ***
> > Patrick wrote:
> >
> > *H3reg^-, zero-grade.
>
> *H3reg^- is full grade, its zero-grade is *H3rg^-. You
apparently
> operate on the assumption that a presumed preverb "*H3o-" will
take
> to ablauting in its position before the verb. That is not a
normal
> behaviour of preverbs. The evasive "preverb o-" is not known
outside
> of Greek, except for the isolated *ó-sd-o-s 'branch'. The
laryngeal
> in *H3reg^- is directly proved by Hittite harganau- 'finger' and
> Avest. &r&zu- adj. 'erect' or m. 'finger', reflecting *H3r.g^-ú-
and
> in part a factitive verb *H3r.g^-n-u- 'make erect, stretch out'.
>
>
> ***
> Patrick writes:
>
> It seems to me that you define things to suit the argument of
the moment.
>
> *H3o is full-grade; *reg^- is full-grade.
>
> The first thing to be noticed is that, in Pokorny, all entries
under *1. reg^- that pertain directly to 'king' and 'rule as king'
have *e: not *e; including Old Indian rá:jati, 'is king, rules'.
>
> The only correct conclusion to draw from this is that the root
for 'king' is *re:g^- not *reg^-.
It is, but a lengthened-grade root segment has weak alternants with
short full vocalism. In EIEC Adams quotes Khotanese forms with
Iranian -a- in rräspu:ra- 'prince', rräysduar- 'princess' as opposed
to kathira:ysa- 'king' from Iran raz- and ra:z- respectively.
Likewise Avestan ra:zar& 'rule, command, regulation' has the gen.
ra:z&:ng, but instrumental ras^na:, also reflecting an ablaut ra:z-
/raz-.
There is also reflex of underlying lengthened-grade in the
participle *H3rég^-to-s with full-rade rather than zero-grade. That
is the same extra mora as in the verb rá:s.t.i 'rules'.
> In no PIE-derived language does 'king' show up with any element
corresponding to an initial *H3o- or *H3-. It is, therefore, not
only unjustified to reconstruct 'king' as *H3reg^-, it is also
simply ridiculous.
No, only Greek would show a reflex of the laryngeal, and Greek has
not retained the words for king corresponding to Lat. re:x, Ved.
rá:j-.
Still, if we draw on all sources of information, we do seem to find
the laryngeal in the Vedic word for king.
Some particples like gha/gha: and nú/nú: are used with length only
when this is required by the metre. The only deviation is gha: in RV
1.178.2 ná gha: rá:jéndra á: dabhan no 'nicht soll uns König Indra
darum bringen' (Geldner); the padapa:t.ha gives the pausa forms
rá:ja: - índrah. - dabhat. In this unique instance the particle
shows a long vowel that cannot be ascribed to metrical concerns. It
is however very easily explained by the lengthening effect of the
following laryngeal in *H3ré:g^-o: 'king'. And the laryngeal is
confirmed by Greek orégo: to the satisfaction of anybody who has not
moved himself into a position of bias against it.
> This is also supported by Sumerian rîg, 'king'.
How is that written? Are you marking vowel length in the word? Could
you be more specific, please?
> So, it is obvious that the root listed by Pokorny as *1. reg^-
should be kept strictly separate from *re:g^-, 'king'.
The opposite is obvious to me.
> As far as the rest of it goes, it is almost a waste to write:
PIE roots are *CVC. A form like *H3reg^- must, therefore, contain
the zero-grade of *H3o- + *reg^-. Whether this *H3o- is the preverb
or not, this is what the first element is in *H3reg^-. Jens
mentioned the form irajyáti. Initial i- is what we expect of an
initial laryngeal in zero-grade in Old Indian.
If it is "therefore", viz. because roots are CVC, you must reserve
two consonants only with a vowel between them to the honour of the
root proper, your statement is entirely circular. You have moved
yourself into a position where you cannot even detect a longer root
when you see one, for your blind spot automatically puts in hyphens
to prevent this.
> And the great brunt of the attested derivations do not show any
trace of *H3(o)- so it is necessary to conclude that it is an added
element to the root *reg^-.
> ***
That is of course because the root is *H3reg^- and only Greek can
generally show the prothetic vowel. Therefore it is in Greek we find
it. We do however have the Vedic forms with iraj- and the
lengthening before rá:ja:, and also Hitt. harganau-. This shows that
the root of the words meaning 'king' and the root meaning 'stretch
out' both had an initial laryngeal. Also the rudimentary ablaut
properties of the root point to identity of the entire set.
Jens