28 Jan 2008

Bronze Age Areal influence in Anatolia and Etruscan

Here's an interesting quote from Calvert Watkins on page 52 of Areal Diffusion and Genetic Inheritance: Problems in Comparative Linguistics (2001) edited by Aikhenvald and Dixon (see link):
"Consider the system of stop consonants. Proto-Indo-European had the traditional three series t d dh; already in Common Anatolian the latter two merged, yielding t and d. The correlation of voice was replaced by one of intensity (tense : lax), with the tense member realized with relative length, thus a tendency to an opposition geminate : simple. Word-finally there was probably since Indo-European times neutralization in favour of the voiced member. But more strikingly it appears that word-initially in Anatolian and there alone among all the Indo-European languages there was neutralization in favour of the unvoiced (tense) member. This explains why when the cuneiform syllabary was borrowed from Semitic, the Semitic voicing oppositions (e.g. TI vs. DI; the capitals denote values of syllabic signs) were ignored in favour of geminate versus simple: word-initial TI or DI to write the same word, but contrasting AT-TI or AD-DI vs. A-TI or A-DI. This system, and the same writing convention, is found in all the cuneiform languages, Indo-European and non-Indo-European alike."
The book goes on to mention that Hittite, Luwian, Palaic and Hurrian all show the same overall typological constraints. That is, they all seem to be pushed towards neutralization in word-initial and word-final positions with a contrast of lax and tense stops being maintained medially. So in initial position all stops tend to become lax (/t/) while in final position they all tend to become tense (such as /t:/ or //, let's say).

Too bad Watkins didn't push this areal convergeance to its furthest logical limit because it unfortunately overlooks the Aegean languages in my view. Other languages in the area at that time which are often ignored in inquiry also, I think, demonstrate a similar pattern. Putting aside Minoan, one of the most important languages of that area during the Bronze Age, surely the Etrusco-Cypriot languages must have been present during that period, centering around the ancient territories of Arzawa and Alashiya (a.k.a. the island of Cyprus). Of the Etrusco-Cypriot grouping I propose, we may easily demonstrate Anatolia's possible influences on Etruscan, if nothing else.

Etruscan shows the same distribution of lax versus tense stops in words. The habits of Etruscan scribes when it came to spelling their language show us that unaspirated stops are the unmarked series while aspirated stops are less frequent and marked. We can see that unaspirated stops are probably "lenis" (i.e. requiring comparatively less articulatory effort) while aspirated stops are "fortis". This assertion fits well with crosslinguistic patterns and it helps to connect Etruscan's phonology with the areal convergeance described for the general area of Anatolia during the early second millenium BCE. Word-initial φ- in Etruscan is quite rare, used for foreign names and words it seems, and the other aspirated consonants are much less frequent in this position when compared to their unaspirated counterparts, as can be seen in the results thus far of my dictionary project. In word-final position, it has been noted that there is a tendency of neutralization of stop contrasts, favouring aspiration (a.k.a. "fortis"). Word-final neutralization towards fortis consonants is precisely what we expect from the patterns we see in world languages. Lenition in such circumstances on the other hand is quite rare (e.g. Lezghian's unusual word-final neutralization of stops to voiced ones). Examples in Etruscan showing word-final neutralization are flanaχ (LL 10.iii) vs. flanac (LL 11.xxix), śeχ (TLE 583) vs. śec (CIE 2611) and zilχ (TLE 126) vs. zilc (TLE 137). These spelling alternations suggest that aspiration is not a contrastive feature word-finally so that Etruscan -c and are both pronounced as /-kʰ/, for example.

This adds to the heap of evidence already showing that the Etruscan language, mythos and culture originated in Anatolia just as Herodotus had wrote millenia ago.


  1. The only thing that could be said about Hittite writing of fortis-lenis distinction is that in writing it looks like all initial consonants are lenis. But we can't actually tell, it's rather hard to write geminates at the beginning of a word with the writing system the had. And they had no signs for aspirates.

    But if there was a fortis-lenis distinction word initially, I guess we would expect Hittite to make use of the voicing distinction the writing system did have. But as I said, it's a bit ambiguous.

  2. Phoenix: "But if there was a fortis-lenis distinction word initially, I guess we would expect Hittite to make use of the voicing distinction the writing system did have."

    If Hittite retained the contrast initially, it would be even more in line with Etruscan. So naturally, being a sly devil, I couldn't possibly object to such a proposal which I may or may not have anticipated a commenter suggesting ;)

    Secondly, if Hittite had lost voicing as a distinctive feature of the language, let's say, while retaining fortis-lenis contrast word-initially, then using voicing distinctions in a foreign system may not have necessarily appealed to the Hittite scribes. A symbol for TI or DI might then have seemed interchangeable to them in initial position.

  3. Normally, I wouldn't toot my own horn quite like this, but I figured you might have some interest in a website/article I did for one of my music classes.
    It's a collection of the different realizations of Hurrian Hymn No. 6, from Ugarit. I've got scores and MIDIs of all 9 known attempts at interpreting the musical notation.
    What I thought might interest you, though, are the three attempted translations of the lyrics. Y'see, they're in a dialectical form of Hurrian, and written by a non-Hurrian-speaker in an idiosyncratic spelling system which doesn't follow the spelling rules of normal Hurrian cuneiform.


    I hope you can forgive me this little vanity...

  4. Neurotically interesting, yes! Even though Hurrian is a little away from my current focus, I still have a place in my heart for the language. I would only recommend that the pages be split up into clearer and shorter categories instead of put into an overwhelming list. But still, the midis are fun to listen to. A lot of work here.