Piotr wrote :
" Word-finally, as in <abe$> and <hame$>, /s^/ is, I think, a
reduced reflex of *c^, not normally found in this position in
Romanian. "
[...]
" There must have been such loans (I also think <hame$> is not
exactly of Balkan Latin age, or it wouldn't have kept its <h> /x/). "
Regarding the romanian h /x/ there are 2 theories in Romanian
linguistic (and not only inside inside Romania - see above Piotr
position on this forum)
1. the 'pro-slavic' theory (I'm not invented here this name, this
is the name used in Romanian books regarding romanian /h/): argue
that the Romanian h /x/ appears in Romanian only during Slavic
influence (so somewhere between sec . VIII- sec. XI).
The main argument of this theory is that there is no 'h' in Later
Latin, so Romanian as a Romance language have no /h/, either. This
theory argue implicitly that the substratum of Romanian Language (and
as result maybe of any language?) cannot introduce its sounds in the
main layer of that language.
For obvious reasons Piotr sustained this theory.
2. the 'pro-dacian' theory (or the /h/ - substratual theory) :
argue that the Romanian h /x/ appears in Romanian as inherited from
the substratum of Romanian language : The Ancient Dacian Language.
For the obvious reasons, I sustained the second theory.
Despite the 'obvious reasons' there are also some arguments too:
As discussed here the presence of h in Dacian glosses is very
probable (++) ( Hydata - toponym, hormia - dacian plant at
Dioskurides etc..), even the number of glosses with 'h' doesn't have
the same frequency as the number of glosses with 'z' (and/or dz) in
the same language (this fact cannot be considered an issue).
A colateral argument for the existance of /h/ in Dacian is that
Albanian 'sk'/sk/->'h'/x/ transformation is older than the Latin
Loans in Albanian, and at that moment we cannot speak about Albanians
but about Dacians etc...
(This collateral argument is based on the assumption (I not discussed
it here, if is true or not) that there is an inheritance link between
Dacian and Albanian too.)
The fact that the Latin didn't have an /h/ at that moment,
explain very well the fact why no Latin word in Romanian have no 'h'.
Both theories agree on this.
But for the 'pro-dacian' adepts, this cannot be an argument to
the assumption : that no other word in Romanian couldn't keep an
inherited /h/ (like we have in (substratual romanian words) : hameS,
hoT, etc...), as 'pro-slavic' adepts say.
Why 'pro-dacian' adepts sustained an 'inherited' h?
First, because the fact that the substratual layer introduced
new sounds in Romanian is fully proved:
a) the existance of Romanian ~a : cas~a 'house'; mas~a 'table'
etc.. (a kind of non-stressed a : like in english 'under' )
b) the existance of Romanian dz /3/ (later passed to /z/) like
in dies -> dzi (but also in substratual words :viezure , mazare,
brad)
c) the existance of sh /s^/ (geusial->guS~a) (already discussed
here, even it appears in a late period, but before Slavic loans )
and 'with your permission' (as Ion Iliescu said when he killed
Ceausescu)
d) romanian /h/ (hameS , hoT)
Next because there are also some H-words too...
Now in order to sustained their 'pro-slavic' theory these linguists
have "something to do" with the word 'hameS' . Why? because ONLY this
Romanian H-word has an Albanian correspondant ( and these scholars
reject the other very probable words like 'hoT' , on the reason that
they don't had an Albanian counter-part).
As result, the real fight is all around the 'hameS' word.
The 'hameS' has all the phonetics necessary to be fully placed
before Romans occupied the Dacia, it has an Albanian counter part,
etc...
But for the 'pro-slavic' scholars their circular reason is the
following :
"we have no substratual words with /h/ in Romanian because the
Balkan Romance didn't have any /h/"
but as regarding the subtratual words , that keep the /h/ ?
"what substratal words, don't tell me about 'hoT' etc...,ah,...
oh, ... 'hameS'? well it should be a loan that arrived later in
Romanian from Albanian, because the Balkan Latin don't have
any /h/ ... '
Up to you to analyze , this type of argument based on the
information I presented above.
Now, how old this 'hameS' coud be? Well if we take a look on the
Toponimy of the Romanian Main Rivers , we found rivers with a
phonetism like :
'Arges^' , 'Mures^' , 'Somes^' , 'Cris^' , 'Aries^' ...
(....please repeat again this list and ...add 'hames^' at the
end. Sound ok, isn't it).
This phonetism is not-at-all a Latin one, but also is not Slavic,
Gothic, Magyar, Cuman, Turks...etc...It is an 'albanoid' phonetism
(see Nis^) that shows that the substratual presence (Dacian) in
Romania was very active, even long before sec IV A.D.
All these toponimy is very old (folowing the SAME phonetism), as
very probable (++) 'hameS' is too.
Best Regards,
marius alexandru