From: dgkilday57
Message: 65622
Date: 2010-01-12
> --- In cybalist@yahoogroups.com, "Torsten" <tgpedersen@> wrote:On second thought, it is not necessary to assume that Langobardic lost the /s/. Instead *-taccare could have been extracted (in the Vulgar Latin dialect of Lombardy ca. 500) from *distaccare or *ex(s)taccare 'to stake out' vel sim. by false division. There are other Romance examples of losing /s/ by this mechanism. And the VL sense of *attaccare could have been 'to impale, drive a stake into, execute (a vampire?) by staking' etc. from which the Old Italian sense would follow by semantic devaluation. No double sense development such as I proposed above would then be necessary.
> > --- In cybalist@yahoogroups.com, "dgkilday57" <dgkilday57@> wrote:
> > >
> > > [...]
> > >
> > > First, regarding (unshifted) Proto-Germanic long tenues, I believe
> > > they pass through Grimm-Verner-Kluge unscathed to remain Common
> > > Germanic long tenues. I have only two examples. The first, Gothic
> > > <atte:kan> 'to touch' (commonly cited in the simplex, which is not
> > > attested)
> >
> > I didn't know that. Do you think the Romance *attak- etc root is related?
>
> No. That one appears to come from Gmc. *stakkan- 'pole, stake, stick' by three or more routes. Gothic *stakka is behind Spanish <estaca>, Frankish *stakka is behind Old French <estache>, and Langobardic *tacca (in my opinion) is behind Old Italian <attaccare>. I do not see the /s/ falling out in Old It. or in Gmc., but in Lgb. The sense development is parallel first to English "stick together", second to "join battle".