Re: [tied] Dacian - /H/ -> seems not possible

On Sun, 07 Dec 2003 08:20:30 +0100, alex <alxmoeller@...> wrote:

>Miguel Carrasquer wrote:
>> On Sun, 07 Dec 2003 00:51:48 +0000, alexandru_mg3
>> <alexandru_mg3@...> wrote:
>>
>>> b) Could we imagine that the Dacian had /H/ ?...I don't think so.
>>> Despite some plant names at Dioskurides (in fact I found only
>>> one : Hormia ) there is no other dacian toponym that contains /H/.
>>> Also romanian loans [sic] from latin don't have /H/ either
>>
>> Latin had no /h/ (it lost it very early on). Therefore, no native
>> Romanian words have /h/, and no substrate /h/'s would have survived
>> among a Romanized population.
>>
>
>which is the logic here Miguel? I see here 2 steps:
>
>1)Latin lost "h" very early thus every Latin word could not have any "h"
>because the "romanised" population _could not know_ where once in Latin
>was an "h".
>
>2) there are substratual words which have had "h" and the "h" was
>preserved.

Not necessarily. Once the population had become Romanized, they didn't
have /h/ anymore in their native language (Eastern Vulgar Latin). The
literate ones among them were perhaps still aware of the fact that H was a
letter to be written, but to be ignored in speaking (the situation in
modern Spanish). In any case, the odds are that if there were any
substratal words with /h/ (and judging by what I know of Thracian, Dacian
and Albanian, it can't have been a very frequent sound), they would have
lost the /h/.

>I gues they are compatible at all. The proof is in the fact that today
>Rom. has "h" and that they _keept_ the Slavic "h" as well, not drooping
>it just because "they are romanised population".

The Slavic "h" is actually a velar fricative /x/, a "stronger" sound than
/h/. Through their extensive contacts with Slavic (probably to the point
of bilingualism), the phoneme /h/ was reintroduced into Romanian. There's
nothing strange about that. As I said, /h/'s come on go. Speakers of
Castilian Spanish went from having /h/ in their substratal language
(assuming it was some kind of Basque), to not having /h/ in Vulgar Latin,
to having /h/ (from Latin /f/), to losing it again, to acquiring [in
certain parts of Spain] a new /h/ (from final /-s/), to losing it again
(with colouring of the final vowel to mark the plural).

=======================
Miguel Carrasquer Vidal
mcv@...