Re: Dacian Dielina

From: Richard Wordingham
Message: 15422
Date: 2002-09-12

--- In cybalist@..., "m_iacomi" <m_iacomi@...> wrote:
> In cybalist@..., alexmoeller@... wrote:
> >> 1. séli:non is also Classical Greek, meaning 'parsley' (once
> >> petroselinum, but renamed Petroselinum sativum).
> Not Petroselinum crispum?
> >> It is the source of the word for 'celery' (A. graveolens) in
> >> English, French, German and Russian at least - probably the
> >> source in most of Europe.
> In Catalan & Spanish we have still descendents of Latin apium:
> api and apio.
> >> 4. Rom. tzelina < Gk. selinon requires an irregular derivation
> >> tz < s

> > [Moeller] is not the same word here. celion is not romanian
> > tzelina. For celion in romanian you have patrunjel how i said
> > before.
> In some European languages, words for celery and parsley are
> related. For instance, Italian word for 'parsley' is prezzemolo,
> coming from Greek petrosélinon (to be recognized in scientifical
> name of the plant) which is a composite pétra + sélinon "stone
> celery", while the word for 'celery' is sedano < selinum (Lat.)
> combined with Greek sélinon. Another one is the Hungarian word.

Parsley and celery are similar enough to have been put in the same
genus (_Apius_), and I am persuaded that the English word 'celery'
does derive from the Greek word. The Latin word 'apius' gave rise to
the English word 'ache', and from 'small' + 'ache' English
derived 'smallage', another name for celery.

> >> 5. Dacian 'dielina' would, I believe, have given *(d)zelina
> >> in Romanian. So in this case, the derivation would require
> >> irregular devoicing and a change in meaning.
> >
> > [Moeller] why irregular? Irregular from what?
> As any normal English reader would realize, a Dacian 'dielina'
> would have evolved not in 'Telina' but in *dzelina > *zelina.
> The fact is Romanian word _is_ /Telina/ => irregular derivation.

I take it 'English reader' = 'reader of English'? Actually, it could
be a sensible question. *If* 'Daco-Thracian' makes sense as a
unitary substrate, the Dacian (Getic?) source was not Dacian as we
know it, but one that had undergone a Grimm's shift like Thracian.
However, returning to Dacian as we know it -

I am assuming that a Dacian /d/ or /ð/ would be identified with
Latin /d/, as least as its functional equivalent, by a Dacian-Latin
bilingual. (I don't think we know for sure that IE *dH > *d; I think
it is conceivable that IE *d > Dacian d, IE *dH > Dacian *ð.
According to ,
Albanian, which *may* be related to Dacian, has IE *d > Albanian
IE *dH > Albanian d - which startled me.) I am confident that there
were Greeks or Romans who associated the Dacian reflexes of PIE *dH
and *d with Greek or Latin /d/. Latin /d/ at the beginning of a word
yields Romanian /d/, /z/ (formerly /dz/) or, possibly, but only very
occasinally, /dZ/ or /Z/.

It would be better if I had access to a list of Dacian loans into
Romanian for comparison. I've read a quote of 230 loans from a sober-
seeming source. (There are a lot of crazy websites around. I have
come across claims that Moldavian and Romanian are Slavonic
languages, and I am dubious about the claim that Dacian and Thracain
were Baltic languages, and I am not referring to Kaulins's twaddle.)

> [Moeller] yeap. The only problem here is if the dacian is to
ask how the greeks should have wrote the "tz" sound. Are there
examples of actual "tz" in any language which are to find ina
greek document?

The date of the Greek matters. "tz" is quite common in Modern Greek,
but of course this isn't what Alex means. I suppose we'd be looking
at Parthian names; I can't think of any other people with affricates
or similar that the Greekes would have dealt with at the time. But
did the Iranain languages have affricates at that time? Actually
there is one other long shot. There's a Coptic letter, normally
rendered 'ti', that was derived from the Egyptian Demotic script
rather than Greek. I wonder if that will tell us anything?

I've recently read of hesitation in Greek between <s> and <z> when
recording Dacian or Thracian (I don't remember which) /z/. It was
dring an unsuccessful Google search for "Dacian vocabulary" (unquoted
in the search specification). Now, Classical Greek <z> was something
like /dz/. I would therefore expect a Greek to record [dz] as <z>.
[ts] would present a problem. I could believe a hesitation between
<s> and <z> again. I suppose <x> (normally [ks]) could be a possible
rendering of [ts].
> >> It would be interesting to know what the words for 'celery'
> >> and 'parsley' are in Albanian, Bulgarian and Hungarian. It's
> >> conceivable that the unetymological spelling with 'c' in
> >> French 'céleri' has modified another language's pronunciation,
> >> and that this change has been passed on to Romanian.

> I couldn't find Albanian word for 'parsley', I'll have to check
> this evening my dictionary. But here are the others (plus Serbo-
> Croatian):
> English parsley celery
> Bulgarian magdanoz celina (voiced /Telina/), kereviz
> Hungarian petrezselyem zeller
> Serbo-Croatian perSun celer (voiced /Teler/)
> Albanian ... (t.b.c.) selino
German petersilie sellerie
French persil céleri
Romanian patrunjel tzelina
Russian petrushka sel'derei

> As it looks from here, a Bulgarian intermediate would fit probably
> better as guess. French céleri has not influenced Balkan words
> celery.

The English, Hungarian, German, French and Russian words
for 'parsley' all come from Greek petro:séli:non. Serbo-Croat and
Romanian seem to have changed the ending, and Russian's re-
interpreted the prefix as 'Peter'.

For 'celery', the Bulgarian and Romanian are the same, but the
question remains - why T- and not s-? The Hungarian's from German,
which in turn is from French, probably the older form in 'sc-'. But
French 's' or 'sc' is /s/, whereas the German 's-' is /z/.

Now, here's a wild thought:

Perhaps German borrowed the word with <sc-> = /ts/ and that affected
some Eastern European pronunciations. German then purged French from
the language - I know that happened, but I can't remember when;
purged <c> from the language - I know that happened; and perhaps
*scellerie became sellerie. That would change the start from /ts/
to /z/. Can anyone comment on this hypothesis?