The Carian language

While advances in the reconstruction of Proto-Indo-European may come a bit slower than they once did, evidentally there are new discoveries to be made in Indo-European linguistics. At the Finnish National Library I came across today Ignacio J. Adiego’s The Carian Language Handbuch der Orientalistik 86 (Leiden: Brill, 2007). It’s a description of Carian, one of the minor Anatolian languages that was soon made obsolete by Greek. Only the discovery of a bilingual inscription in 1996 finally settled the issue of decipherment of the Carian script.

The bulk of the book is the long story of the decipherment of the language and the various sources. Me, I’m not too interested in decipherment issues, so I turned straight to the survey of the language. Though attestations are exceedingly brief and fragmentary, Carian is squarely an Anatolian Indo-European language. It maintains the PIE laryngeal *h₂ like the other Anatolian languages, with the exception of Lydian. Interestingly, the laryngeal becomes a tectal stop in Carian instead of a (velar?) fricative in Hittite and Luwian. Could this serve to more securely identify h₂ as [x]? What sets Carian in the Luwic group of Anatolian languages is the palatalization of PIE *k > s, while the rest of Anatolian has k. The nominal morphology of Carian shows some general IE features. The acc. sg. ending is -n, and the acc. pl. ending is believed to come from *-ns. Unfortunately, there are not sufficient examples of verbal morphology.

Some Anatolian reading

a hettita a hettita
különös nép a hettita
azt hiszi minden hettita
mindenkiről hogy hettita

Sándor Weöres, “Le Journal
(from Csontváry-Vásznak, 1953).

Recently I’ve read two papers about an IE branch I rarely do much with, Anatolian. The first item is Jaan Puhvel’s paper “Whence the Hittite, Whither the Jonesian Vision?” in Sprung from Some Common Source: Investigations into the Prehistory of Languages, ed. Lamb and Mitchell (Stanford University Press, 1991), one of the couple of worthwhile papers in the whole sorry volume. Puhvel’s purpose is to show Indo-Europeanists that Hittite really is an object worthy of study, not some curiosity that we can avoid. He gives some examples of how quintessentially Indo-European it is, trying to put to rest Cowgill-ish notions that Anatolia split off so early that we can limit our investigations to the other languages and ignore Hittite. He ends with the plea that scholars not only enjoy Hittite, but even be able to make use of primary sources so that they don’t depend on the same edited treatments again and again.

It’s an entertaining paper and worth reading for anyone interested in comparative Indo-European linguistics. The only thing stopping me from taking his advice to heart is the lack of a good primer for Hittite. There’s apparently nothing out there as easy and pleasant to use as Beginner’s Lithuanian by Dambriunas, Schmalstieg, and Klimas, or Handbook of Old Church Slavonic by Nandriș and Auty. No, Beginning Hittite by Held, Schmalstieg, and Gertz doesn’t count. It’s too short, unfriendly, and dreadfully typeset. The online course I recently mentioned is a good start, but I can’t read it in the train, and that’s where half of my study goes on.

Calvert Watkins contributes “An Indo-European Linguistic Area and its Characteristics: Ancient Anatolia. Areal Diffusion as a Challenge to the Comparative Method?” to the volume Areal Diffusion and Genetic Inheritance: Problems in Comparative Linguistics, ed. Alexandra Y. Aikhenvald and R. M. W. Dixon (Oxford University Press, 2001). He makes some intriguing comments about areal convergence as a way to explain several phonological changes in this part of the world. The first is the adaptation of Proto-Indo-European’s phonology to a general Anatolian “portmanteau inventory” of phonemes, which all languages in these parts came to hold regardless of genetic origin:

p t ʦ k () [+ tense, + long]
b d g () [- tense, – long]
(f) s H [+ tense, + long]
(v) z h [- tense, – long]
m n [+/- long]
l r [+/- long]
w y

The second shift that Watkins attributes to areal convergence comes two thousand years later. The voicing of unvoiced stops after nasals in early modern Greek follows the dying Anatolian languages in western Asia Minor. The examples he gives include the spelling in Lycian of the Greek name Δημολ[ει]δης as Ñtemuxlida, and Greek dialect πεδε ‘fifteen’ in the region where Anatolian Sidetic is found (everywhere else πεντε). I’m sceptical about this, as I like to ascribe the changes resulting in modern Greek to as recent an era as possible, and so this would be when those Anatolian languages were already long gone.