>"But this would - at least IMHO - be a hint that the word is a
>loanword from Hungarian."
>
>In order that your arguments become credible you have to
>underline first a single fact: What is the etymology of
>Hungarian word 'szidni'?
No, this kind of judgment is flawed. Since, if it can be proved
that a word X in the language C is a loanie from the language
B, then it doesn't matter that it is a loanword in the language
B from a language A. (In our case, it doesn't matter for the
Romanian language whether Hungarians once, in the Middle Ages
had been fond of that word upon hearing it in its initial
(let's say) Slavic form. I'd rather assume Slavic, since
Hungarians contacts with Albanians were scarce as compared those
of South-Slaves with the same Albanians.)
>But without a clear origin of Hungarian word 'szidni' your have
>not enough credibility to sustain and go on with your claim...
I reiterate the above in other words: if science (i.e. based on
scientific criteria) has come to the conclusion that a word
in a language is a loanword taken from another language, then
it does not matter for the analysis and discussion that the
other language itself had gotten that word from a... 3rd
language, and this one from a... 4th one (and so forth).
The other way around: if we could detect traces of some
substrate (PIE, Thracian or whatever) in the Romanian <sudui/re>
and <sudalma> which could thwart all judgments based on comparisons
with the Hungarian <szid/ni> and <szidalom>, then we could start
scratching our heads pondering over some sort of a revision of
the prevalent conclusion. Until then, we read it in the
dictionary whenever we open it. And, by and large, professional
lexicographers tend to include in their dictionaries what's
up-to-date in linguistics and not things that average lay people,
native-speakers or not, in the streets, deem suitable.
There are many more Magyarisms in the Romanian language.
(Moreover, there are some Hungarian loanwords in everyday's
Romanian language spoken *outside* Transylvania, Banate
and Maramuresh, namely loanwords that are not used by or
unknown to Transylvanians, Banatians and Maramureshers
themselves; e.g. borviz "mineral water", for which in Hungary
they anyway usually say <aSvány víz>) And no wonder:
Hungarians (qua their apostolic kingdom) had a strong impact
in Eastern Europe.
As for swearing/cursing, I'd also pay heed to the fact
that the genital-anal-fecal-blasphemous kind of talk in
the Hungarian-Romanian-Slavic-Romany (or Gypsy) area is one
of the strongest-ugliest in the world. (AFAIR, a Munich
linguist, Asam, wrote a book referring to this as well;
methnks in the 1980s. I haven't read it myself, but I was
born and raised up in the... eye of this linguistic...
cyclon, so I'd be inclined to say the author is right. :))
>we obtain the PAlb *sadunja.
Even if one would prove that this one was the origin of
the Hungarian <szid-> (say, via a Slavic "path"), you
cannot take it for granted that Romanian <sudui/re> +
<sudalma> stems from it *directly* (i.e. not via the
Hungarian link).
By the same token, there are lots of Romanian lexems from
Latin via French, German and even Russian: these cannot be
taken into consideration whenever we analyse the kind
of the Romanian language as one amongst Romance languages
(i.e. transformations of the Latin language or Italic idioms).
In order to be relevant within this etymological discussion,
such a word must stand in a *direct* relationship with the
idiom considered to be the "donor" or to be in a kinship
relation as "siblings". Take Albanian: such words in direct
relationship, i.e. without the mediation of other languages
of the linguistic region, are <mo$ & moa$ã; mânz; mazãre;
viezure; zãr; cãciulã; copil; cioc> etc.
>'the lost of intervocalic d preceed the au>a transformation
>in Albanian'
Prior to dealing with ancient, medieval and modern Albanian,
I'd prefer to pay heed to the inner "mechanisms" of the
Romanian language.
>Strange enough in Romanian we have a verb 'sudui' that can also
>be simple derived to a Proto Romanian form *sudunja.
Then I'd expect to have the verb *a suduni in some regions,
esp. in the Banate region, where [n] hasn't vanished up today,
in contrast with standard Romanian and the Romanian subdialects
all over the territory (e.g. panRomanian <tu spui> "you say",
but in these small areas they say <tu spuñi>!). Is there in
the Romanian-speaking area any subdialectal spot that uses
the word with a preserved [n] in it?
Another aspect we must pay attention to: the IV-th conjugation
in Romanian is... over-inflationarily used for verbs adopted
as loanwords from other languages. This tendency has remained
utterly strong in any Romanian native-speaker until today, and
it'll stay such many decades to come.
So, if I were shown a reasonable path from *sudunja to <suduí>,
as well as a plausible explanation for the genesis of *sudunja,
I'd say "Well, the Hung. <szid> etymology seems rickety."
But I'd also need an interpretation revision for <sudalma>;
i.e. I'd need something stronger to explain it than what's
been given (even to a non-linguist) by the Hungarian <szidalom>,
which differs only in the root vowel, whereas the suffix is
the same [the Hungarian part changing its form depending on
contexts: alom <-> alma, elem <-> elme; an example for the
latter, since I didn't offer one in my previous post:
<szerelem> "love", but <szerelmeS> "in love; inamorato/-a"].
That is: the particle [vowel+L+vowel+M] changes to
[vowel+LM+vowel], back and forth, because of certain morphologic
and syntactic necessities pertaining to the Hungarian language;
whereas -alma in Romanian can be found only in <sudalma>, you
said - and I'm afraid I'll tend to agree with you in this respect.
Rare similar phonetic occurrences such as <palmã> and <calm,-ã>
are of no use, since these aren't suffixes, and, on top of that,
<calm,-ã> is a neologism in Romanian (i.e. it isn't inherited,
AFAIK).
>But among all these examples the most solid one that I found
>is in my opinion: *baruk ; *baruk-ta for Alb. bark and
>Rom. burta 'belly' (PIE *bher-1 'to carry')
Really? But what about [k] > [t]? Why couldn't stay [k] in
Romanian as well? Average Romanian phonetics skills enable
any native-speaker to distinguish between [k] and [t] (except
people who are born with the malformation of the palate, ie.
a split palate). Are you aware of similar [k] > [t] transformation
patterns in Romanian? (I mean: the genuine velar [k] and the
genuine dental (even alveolar) [t]!)
Moreover, if PIE *bher- were to be considered as the root for
Romanian <burtã> "belly, stomach", then would it be a 'conditio
sine qua non' that it went through the Albanian <bark>? (A
vanishing of -uk- together with [a] > [o] or [&] in *baruk-ta
could be a tad more plausible for the emergence of a Romanian
<burtã>. But I don't know whether these assumptions are
warranted at all.)
> Marius
George
PS: If there is no scientific approach to that, then, to me,
it's like it's said in a song sung by Ella Fitzgerald: "It
don't mean a thing, if it ain't got that swing!" <doo-wat,
doo-wat, doo-wat> :-)