--- In
cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:
>
> On 2007-10-16 18:21, alexandru_mg3 wrote:
>
> > Can 'hat' be traced back to a PIE *kaT-T- , as is the case of
Latin
> > cassis and PAlb. /kac^ula/ ?
>
> I wonder if MLat. casula 'cloak with a cowl' (hence kos(^)ulja or
the
> like for 'shirt' in various Slavic languages) belongs here rather
than
> being a diminutive of <casa>. I mean it could have been a
substratal
> word of the Balkan area borrowed into Vulgar Latin, perhaps
with 'hood'
> as the original meaning.
But Romanian /k&c^ul&/ 'hat' with c^ cannot be from Latin casula that
is a separate word (the single s-rules are : Latin s > Romanian s;
Latin sj > Romanian sh written s, )
Because the c^ of Romanian /k&c^ul&/ 'hat' cannot be originated on
the s of MLat casula , /k&c^ul&/ cannot be 'derived' from MLat casula
only the invers relation could be true...
But maybe your point was related only to MLat <casula>?
Having a Subtratual /k&c^-ul&/ from *kaT-tó- (with a direct cognate
Latin cassis 'helmet' and an indirect one *kaT-nó-
(Germanic "hat")) : we have in /k&c^ul&/ 'the live output' of the
Dental+Dental in Proto-Albanian...If true, this is quite amazing...at
least for me.
> I'd put my money on a Klugean reconstruction of
> the Gmc. "hat" word as *kaT-nó-, in which case *kaT-tó- would be
related
> but not quite the same.
>
> Piotr
>
If so, I will put my money too, on the same 'color' with you...
(even to be honest here: my first trial would have been *kaT-T-nó- ,
trying to preserved *kaT-T- anywhere, and for this reason I have
asked for Kluge's Law timeframe, trying to find out if *-T-T-nó-
cluster can give -tt- in Germanic)
Thanks, again.
Marius