Re: [tied] Re: Alb. =?UNKNOWN?Q?af=EBr/afro/afroj?=

21-12-03 21:17, Abdullah Konushevci wrote:

> I gues that Alb. <afër> 'near, close to' is formed by zero-grade
> combining form *n. > Alb. /a/ 'not' and *(s)per-1 (*sp > f) with a
> wide range of extended senses, like "beyond", "far". So, orignary,
> the meaning of Alb. <afër>, derived from *n.-sper was 'something
> which is not far". <afroj> and <afro> (cf. lat. approximus) are
> derivatives of <afër>, as <afri> 'closeness',
> <përafërsisht> "approximately", etj.

I agree. There's no reason to deny that the word is inherited, and if it
is, the /f/ can only reflect *sp. It's more difficult to identify the
root involved. It might be *sper-, as you suggest, but I'd also consider
*speh1(i)- 'extend, stretch' hence 'grow, prosper, thrive', cf. Slavic
*sporU, Skt. spHira- 'large, abundant' < *sp&1-ro-.

> Extended form of *n.- 'not' I see in negation prefix <as> 'not',
> derived from *n.-s (cf. also <mos> 'don't' < *me:-s), another
> argument of outcome of PIE /*s/ > Alb. /s/, besides /*s/ > /gj/
> and /*s/ > /sh/.

Why do you insist on this unlikely outcome? The final /s/ of <mos> is
generally regarded, with good reason, as reflecting enclitic *-kWe
(*meh1-kWe 'and don't ...'). This reconstruction is far more convincing
than the entirely ad hoc assumption of *-s (what would it have been or
meant?).

Piotr