Re: Kluge's Law in Italic? (was: Volcae and Volsci)

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

>
> --- In cybalist@yahoogroups.com, "dgkilday57" <dgkilday57@> wrote:
> >
> > --- In cybalist@yahoogroups.com, "stlatos" <stlatos@> wrote:
> > >
> > > --- In cybalist@yahoogroups.com, "dgkilday57" <dgkilday57@> wrote:
> > > >
> > > > --- In cybalist@yahoogroups.com, "stlatos" <stlatos@> wrote:
> > > > >
> > > > > I don't think any additional complications are needed. It's already seen that tx > tHx was opt.:
> > > > >
> > > > > pathana- = broad Av; [e>a-a] patáne: = flat dish G;
> > > > >
> > > > > and tH > T > D > d in Latin.
> > > >
> > > > No; see <status>.
> > >
> > > Are you saying that had -tH- in PIE and tH>t not tH>T in Latin? If not you might be arguing against t>tH by x as I wrote above. If so, that would be ev. if the rules were reg., but they're not.
> >
> > What I am saying is that PIE *h4 regularly aspirates preceding Skt. -t-, and produces Av. -T- unless -s- precedes the dental; Av. -st- corresponds to Skt. -sth- in that case. But Italic stops are not aspirated or fricativized by *h4.
>
> You'd need at least h4 to h12 to make sense of aspiration w/o opt. changes.
>
> What are the sounds that you have for h2 and h4, if both apparently change e>a and one causes asp. before and the other after in some IE? What happened in pairs like:
>
> mitrá- = friend, mithuna- = a pair S;
>
> pri:tá- = pleased S; fritha- Av;
>
> or
>
> ti:rthá-m = passage/way/ford V S; tHúht = ford Khow; túr = animal trail Kamv; tìltas = bridge Lith; etc., with PIE * tLxàtló+ tL-x-t.r.+ showing t-t > t-tH / tH-t / t-? even w/in Indo-Iranian?
> >
> > > Instead of repeating my many arguments in favor of that, I'll say: yes; see stabulum L; staflar- U; what part of "optional" don't you understand?
> >
> > Those Italic words reflect PIE *sth4-dHlo-, and no optional soundlaws are needed.
> >
> There's no ev. for any *-dHlo- instead of *-tlo- anywhere. In Greek and Italic it seems almost impossible that they would follow so many V (that is, older (V)x which caused t>tH, etc.) instead of having random distribution for stem-endings. For ex., why are tafle (loc) U; tabula L; the same? Where are the *-tlo- endings after -a- < -x-?

A deverbative stem in *-dHlo-/*-dHleh2- is supported by Greek <génethlon> (Aesch.), <genéthle:> (Hom.) 'birth, origin, descent'. Like this word, Latin <pa:bulum> 'food, fodder, nutriment' and <fa:bula> 'talk, tale, story' have full grade, but <stabulum> 'abode, den, stall' has zero-grade of the root, as does <tabula>. The Umbrian <tafla> (loc. sg. <tafle>, Tab. Iguv. IIb:12) was "evidently a broad flat carrier of some kind on which [were] to be placed not only the fire but the [cut-off parts] of the victim as well" (Poultney, Bronze Tables of Iguvium, p. 195). The old suggestion that this word comes from *telh2- 'lift, raise' by dissimilation of *tl@-dHla: (ib. 91, questionably) can be discarded, since zero-grade would have given Italic *tla:- as in Lat. <la:tus> 'carried'. Most likely the root is *teh2- or *teh4- 'take away, remove, steal' observed in the Dvenos prohibition NE MED MALOS TATOD 'a bad man shall not remove me'; Old Church Slavic <tajiti> 'to conceal', <tatI> 'thief'; Proto-Celtic *ta:tant- > Old Irish <táid> 'thief'.

In my opinion Lithuanian <tìltas> 'bridge' and Sanskrit <ti:rthám> 'ford, watering-place' reflect *tlh4-tó- from *telh4- 'cross over'. Skt. <tí:ryati> 'crosses over' and <tí:ram> 'river-bank' might belong here, or could equally well be grouped with Skt. <tárati> 'crosses over' from *terh2-, a different root.

Another example of -:th- < *-h4t- is probably Skt. <gu:tha-> 'filth', Avestan <gu:þa-> 'id.', against Skt. <gu:nam> 'excrement', evidently from a zero-grade *guh4- or *gWuh4- seen also in <guváti> 'defecates'. If related, Old English <cwéad> 'filth' and Old Frisian <qua:d> 'evil, bad' must continue *gWóh4u-dHo-; likewise OIr <buaidir>, Middle Welsh <budr> 'dirty' < PCelt *bowdro- < *gWóh4u-dHro-. In this view the original root was *gWeh4u-, with laryngeal metathesis regularly occurring in the Indo-Iranian zero-grade *g(W)uh4-. This is what we have from *bHeh2u- 'be', Skt. aor. impv. <bodhí> from *bHéh2u later enlarged by *-dHí, metathetic zero-grade *bHuh2- in aor. ind. <ábhu:t> 'was', secondary full grade *bHewh2- in <bhávitum> 'to be', pres. ind. <bhávati> 'is'.

In my opinion Skt. <yu:thá-> 'flock, herd, crowd' cannot belong to the same root as <yu:tís.> 'combination'. The latter belongs with <yu:nam> 'band, cord', <yuváti> 'binds together' from *jeu-h2- 'join completely, mix, blend' while the former represents *jeh4u- 'gather, collect' vel sim.

If Skt. <mitrá-> 'friend' is related to <mithuná-> 'paired', either InIr *-thr- has been regularly reduced to -tr- in Skt., or PIE *-th4r- was already reduced to *-tr- in Indo-Iranian. Avestan has -þr- in such words in either case. Av. <frita-> is what I have seen cited as cognate to Skt. <pri:tá->, with no discrepancy. An actual discrepancy does occur between Skt. <prathamá-> 'first' and Av. <frat@...>, Old Persian <fratama->, Middle Pers. <fratam>. I suspect that the aspiration in the Skt. word is due to contamination with <pr.s.thá-> 'standing over, upper', which has similar superlative force although differently formed. No such discrepancy occurs between Skt. <pratará-> 'further, future' and Av. <fratara-> 'more in front, higher'.

I am not sure what you are asking in your last question about *-tlo-, but I suggest you look at Latin nouns in -a:culum.

As for the sounds of the laryngeals, my working hypothesis is that they were approximants which under certain conditions narrowed down to fricatives. I suspect that *h3 was originally uvular, *h2 palatovelar, *h4 cerebrodental, and *h1 labial, but I am open to reasonable objections.

DGK