From: Miguel Carrasquer
Message: 18272
Date: 2003-01-29
>I have no quarrel with that, it's the prudent thing to think. In my
>----- Original Message -----
>From: "Miguel Carrasquer" <mcv@...>
>To: <cybalist@yahoogroups.com>
>Sent: Wednesday, January 29, 2003 12:41 AM
>Subject: Re: [tied] Laryngeal theory as an unnatural
>
>
>> One of the strongest cases we have for *-s is in the word *dom, *dem-s, and given the structural similarity of *pod-/*ped-, I think reconstruction of a G. *ped-s is fully justified, even if unprovable.
>
>I accept *dem-s, since to do otherwise would mean to ignore the evidence. I agree the case for *nekWt-s is quite convincing. *-ei-s and *-ou-s/*-eu-s genitives are of course entirely uncontroversial. But to insist that *ped-s is _fully_ justified given the structural similarity of *pVd- and *dVm- is, to my cautious mind, a bit like claiming that <house> must have had an umlauted plural like <mouse> (extra evidence: Christ Church, Oxford, is sometimes locally refered to as "The Hice" ;-)). It means placing one's own theoretic taste above the evidence. From the point of view of safe methodology, *pó:ts/*ped-ós should be favoured (at least for the time being) as the pattern consistent with the available data and accounting for the known variants.