Page 1 of 1

Croatian Toponyms

Posted: Tue Jan 16, 2018 8:14 pm UTC
by FlatAssembler
So, what do you guys here think about my alternative interpretation of the Croatian Toponyms?
I don't think that knowledge of the Croatian language is required to evaluate my theories, but the knowledge of linguistics is.

In short, although Croatian is a Slavic language, I think there aren't that many Slavic toponyms in Croatia, because, if there were, they would make sense to those who speak Croatian, and most of them don't. So, I try to reconstruct the language of the ancient Croatian toponyms (the Illyrian language) and explain the toponyms using it. I mostly refer to Indo-European roots (and suppose a series of sound changes that occurred from Proto-Indo-European to Illyrian), though I explain a few toponyms by claiming that some repeating elements (with definite meanings) were borrowed into Illyrian from a Pre-Indo-European language.

Re: Croatian Toponyms

Posted: Wed Jan 17, 2018 12:58 am UTC
It's not at all surprising that Croatia is full of non-Slavic toponyms seeing as the Croats migrated there during the Avar Khaganate c. 600AD. For comparison, the Anglo-Saxons migrations started only about a century earlier and an awful lot of toponyms are Celtic (e.g. York, from Eboracon, any of the rivers Avon, or the Thames), or Latin (e.g. Chester, Eccles from ecclesia, and Stratton from strata), as well as a few possible pre-Celtic ones (e.g. the river Granta, now mostly called the river Cam) as well as lots of later Norse (e.g. Grimsby, Scunthorpe, Derby), and Norman (e.g. Stoke Mandeville, Beaulieu, or the second element of Chesham-bois) toponyms. It's also worth noting that even lots of the native English toponyms do not have a transparent etymology (e.g. Stevenage is etymologically stiff-oak but you'd struggle to know that) so some particularly early Slavic toponyms may not be transparent.

With that in mind, one issue that grabs me immediately is a few of your toponyms are not strictly of modern Croatia and so aren't really relevant to any attempt to reconstruct Illyrian. In particular, Danube is first recorded in Greek sources focussing on its mouth into the Black Sea and it's well agreed to be a Scythian toponym closely related to the Dnieper, Dniester, and Don; Croat is also used to describe the Slavic inhabitants of northern Illyria almost as soon as the Croats arrive, this would be surprising if it were a toponym attached to the territory, but entirely expected if it was just a demonym they brought with them (its etymology is very opaque though, although usually assumed to be Scythian in origin).

Ok, now onto the sound changes you propose.

Merging all three unvoiced velar series (kj, k, and kw) is attested only in Tocharian (which also merges the voiced velars, whilst you keep gj distinct; this inconsistency is unlikely).

kj>j is attested only in English and through a very different process taking it via ȝ. It is hard to see how m could condition this change particularly if it also has to disappear afterwards.

You've got bh>p but dh>d which is unattested and unlikely.

Your other sound changes seem broadly plausible albeit a lot more specific than I'd expect. I don't know the conventionally accepted etymologies of most of the toponyms you look at, but your proposed Illyrian seems to involved a few too many implausibilities and is based on evidence of unclear relevance. If you want to actually attempt a reconstruction of Illyrian, I suggest you start with vocabulary accepted to be Illyrian and then see if any of your conclusions apply more generally rather than going vice versa.

Re: Croatian Toponyms

Posted: Wed Jan 17, 2018 8:42 pm UTC
by FlatAssembler
Thanks for your reply!

Some particularly early Slavic toponyms may not be transparent.

Sure. But trying to explain Croatian toponyms without an obvious Croatian etymology using Croatian (Slavic) roots would probably just get me into the Bongo-Bongo Effect. I simply know Croatian too well to be objective.
You've got bh>p but dh>d which is unattested and unlikely.

Well, I'm aware of that. However, as far as I can see, there appears to be no other way to explain the element *puk~*puh, which appears in quite a few Croatian hydronyms (if not from PIE *bhogj).
As for dh>d, well, the primary reason for that supposition is the hydromyms containing the element *da (Danube, An-dautonia, Darna…). I supposed it came from *dheh2 (to flow). However, not all etymological dictionaries agree that the 'd' was aspirated. Pokorny, for instance, reconstructs the root simply as *da: ... ary/dā.htm
The reason it's supposed to be aspirated is the connection to the Latin word 'fons'.
The toponym "Dalmatia" is connected, by an old tradition, to the Albanian word "dele", meaning "sheep" (from PIE *dhel). I accepted that tradition, although I don't really see why would that be more likely than the connection to the Indo-European root *del (wide). Descriptive toponyms are more common than the toponyms named after animals.

Re: Croatian Toponyms

Posted: Sun Jul 22, 2018 7:56 am UTC
by FlatAssembler
In case somebody here is interested, I've contacted Dubravka Ivsic, perhaps the most prominent Croatian linguist, via e-mail and published her responses on the web-page. Just click the "show/hide details" links.