Re: Mapping the Origins and Expansion of the Indo-European Language

--- In cybalist@yahoogroups.com, "dgkilday57" <dgkilday57@...> wrote:

> --- In cybalist@yahoogroups.com, "Francesco Brighenti" <frabrig@>
> wrote:
> >
> > d in Vedic Sanskrit adbhis, adbhyas has long been thought to
> arise through assimilation to the following voiced consonant (*ap-
> bhis > *ab-bhis > adbhis). An analogous dissimilation pattern is
> also seen at work in the dat.pl. of Vedic Sanskrit napa:t-
> 'grandson', i.e. nadbhyas (< *napt-bhyas).
>
> But that is not quite parallel. Since the zero-grade of 'foot' in
> compounds, *-pd-, becomes -bd-, we should expect *napt-bhyas to
> become *nabdbhyas first. The attested form <nadbhyas> would then
> result from loss of -b- through cluster simplification, not
> dissimilation.

This problem was already discussed by, among others, linguist Johannes Schmidt (in a 1895 work) -- see pp. 59ff. at

http://home.us.archive.org/details/kritikdersonante00schmuoft

> I have seen several other attempts to explain the adbhis-adbhyas
> problem going back to K.F. Johannsen, and have found none of them
> convincing, so I think root-suppletion is a valid option here.
> After all, we do have this root in Avestan <aDu->.

O.k., now I see (after a search on the Web) that this *ad- root for 'water' has been proposed long ago -- see, e.g. here:

http://tinyurl.com/cnrqt6u

<< Die ai. I. pl. und D. pl. <adbhis>, <adbhyas> haben einen Stamm *ad- 'Wasser' zur Voraussetzung, der zu aw. <adu> 'Wasser, Bach', lit. <od-menis> 'Flußmündung', gr. <adis-> < *ad-s- 'Schlamm', alat. <assyr>, jünger <asser>, <aser> 'Blut' gehört, welch letzteres wegen /ss/, /s/ < /ss/ nicht zu ai. <asan-> gestellt werdern kann. >>

Is it an ablaut variant of *ud- (= *uod-/*wed-) as is maintained by A.G.E. Speirs here:

http://tinyurl.com/buzk8ao

Otherwise, how many similar but discrete IE roots for water do we get? (*ap-, *ab-, *ud-, *ad-,...)

Best,
Francesco