Re: [tied] Latin testu:do and tortuca

On 04-12-28 17:33, Joao wrote:

> Does anybody know the etymology of Latin testu:do: (-inis) "turtle" ? It
> seems to be related to testa, but what's the analysis? What is the
> meaning of the suffix.

Itr's from the u-stem <testu> 'vessel, earthen pot, lid', related to
<testa>, hence *testu:tus 'shelled, having a shell', like <cornu:tus>
'horned' from <cornu>. The voicing of *-t before the -o:/-in- suffix is
a regular affair in Latin, one of the facts presented by Olsen as
support for initial *h3- in the "Hoffmann suffix" (*-h3on(h2)-,
according to her analysis).

> Neo-Latin derivatives, tortuga, tortue, and English tortoise and turtle
> came from a Latin *tortuca? What's its origin? Portuguese has tartaruga,
> which usually analysed as corruption of tortuga, or from an odd Greek
> tartaroukhos "Hells's Guardian". What is the current view.

Then same. It remains a valid etymology. <tartaroukHos> 'controlling
Tartarus' --> Lat. tartaru:chus, fem. -a > Late Latin tortúca etc.
(presumably contaminated with <tortus> 'twisted').

> Portuguese has a name for terrestrial turtles, ca´gado (in Brazil we
> also use jabuti, of Amerindian origin), which would come from Latin
> *cacitus or *cacabatus. Any idea?

I think it's a good idea, though I'm really out of my depth here.
<ca:cabus ~ caccabus> 'cooking pot' (<-- Gk. ka(k)kabos 'three-legged
pot') and <ca:caba:tus> are attested Latin words. The latter's attested
meaning is 'black, sooty (like a pot)', but it may well have had a sense
parallel to *testu:tus.

Piotr