Anatolian Spring in Copenhagen

It is well-known that the cuneiform script that the Hittites adopted from Mesopotamia contains different signs for voiceless and voiced stops (e.g. TA vs. DA, KA vs. GA, KI vs. GI, etc.). Moreover, it is well known that Hittite possessed two series of stops, namely a fortis series (reflecting PIE voiceless stops) and a lenis series (reflecting PIE voiced and voiced aspirated stops). In most handbooks it is stated that in word-initial position, however, the original distinction between fortis and lenis stops was lost in Hittite, and the two merged into the fortis series. Moreover, it is stated that in spelling, the choice between signs for the voiceless stop (e.g. TA, KA, KI) and signs for the voiced stop (e.g. DA, GA, GI) does not correspond in any way to the etymological value of that stop, and is in principle random: if there seems to be any pattern, that pattern is caused by scribal convention or just by fashion.

In this talk I want to review the distribution of signs for voiceless stops vs. signs for voiced stops in word-initial position in order to research if there really is no meaningful distribution between them that corresponds to a phonological distinction.

Bemærk at denne forelæsning afholdes på tysk.

It has been suggested that in Luvian dental stop in word initial position partly stems from Proto-Anatolian *sH-, e.g. dūr ‘urine’ from *sh₂ur, cf. Hittite sēhur. This development will be treated in detail, discussing ways of possible phonological explanations.

Besides, an attempt will be made to show a phonologically parallel development in PIE itself. Three samples will be used for this purpose: late PIE *dak̂rú- (and *ak̂rú-) ‘tear’; *daiwér- ‘brother of husband’; and *dou-s- ‘upper arm’.

Hittite primary u-stems are said to belong to an old stratum of the derivational history of PIE and to show a variety of archaic features. Nevertheless, there are also several secondary u-stems based both on athematic and thematic stems, which are of some interest and which are presented and analysed in the paper presented.

One important implication of the demonstration by Kloekhorst (2006) that the Hittite verb ‘to overcome, conquer’ is only /tarx^w-/ (no stem †tarḫ- exists) is that some instances of medial *h₃ do appear in Luvian and Hittite as -ḫ-, thus refuting the widespread view that all instances of medial *h₃ were subject either to deletion or assimilation. This new finding calls for a complete reassessment of the fate of medial *h₃. I will review all the relevant evidence to try to determine the conditions for its retention or loss/assimilation.

Most scholars nowadays assume that in Hittite there is a correlation between plene spelling and the place of the accent. Yet, details are far from clear. In this talk I will present one of the outcomes of my current research project about accentuation in Hittite, which will deal with the plene spelling of the vowel -i- and the accentuation of dat.-loc.sg. forms.

Bemærk at denne forelæsning afholdes på tysk.

In Old Hittite the collective was still preserved as productive category functioning not only as plural of singular neuters but also of singular nouns with common gender. Concerning the prehistory of the Hittite collective, several problems and questions will be discussed:

What was the oldest meaning of the PIE type *u̯edōr (cf. Hitt. widār ‘water(s)’)?
Why are some collective-like looking Hittite nouns like zashai- ‘sleep, dream’ common gender?
Are Hitt. ishunau- ‘upper arm’ and harganau- ‘palm (of hand), sole (of foot)’ original collectives?
How productive were i-collectives in PIE and Anatolian?

The language of the Hittite rituals seems to be simple and of little interest. It can, however, be shown that the Hittite technical language is the result of a long tradition.

Recent research, most notably by Starke, Rieken, Rößle, and Widmer, has radically altered our picture of i-stems in Anatolian. Starke’s discovery of ‘i-mutation’ in the western Anatolian languages has refuted the notion that i-stems were rampantly productive in those languages. Rieken has shown that ‘i-mutation’ has nothing to do with either of the feminine ‘motion’ suffixes of Core PIE, while Widmer has made a convincing argument that the vṛkī suffix is present in Anatolian, but not in the function of marking the feminine. Rößle has established that oxtyone i-stem nouns were secondarily incorporated into the diphthongal stems in -āi-. No synthesis incorporating these new insights has yet been made, nor have certain still outstanding questions been addressed. I will seek to provide at least a provisional coherent overview of our current understanding of i-stems in Anatolian.

From the reviews of my Etymological Dictionary of the Hittite Inherited Lexicon that I have thus far seen, it appears that one of the more controversial aspects of my views on Hittite phonology is my postulation of a synchronic phoneme /ʔ/ (glottal stop). If I have not been able to convince colleagues, I apparently have not presented my arguments clearly enough. In this talk I therefore want to discuss the relevant material again in a more elaborate way, pinpointing what the exact points are that in my view force us to postulate a phonemic /ʔ/.

Bemærk at denne forelæsning afholdes på tysk.

In this paper I shall try to present samples of the different stages of research in Anatolian word formation:

1. There is Cuneiform Luvian ārrazza/i- of still unknown meaning. While discussing possible meaning and morphology I shall suggest an etymological connection with the PIE root *h₁erĝ^h- that was reconstructed by Watkins 1994b (1975).
2. In the case of Hitt. harsāwar ‘tillage, broken ground’ discussion can start from a more developed level: this word is a product of secondary derivation like e.g. Vedic maghavan- ‘generous’.
3. The ablaut of the typologically oldest Hittite reduplicated nouns like memal ‘groats’ cannot be investigated without a glimpse to nouns of other IE languages, e.g. YAv. baβra- ‘beaver’ and Vedic jígarti- ‘devourer’, cákri- ‘doing’ and, last but not least, to the reduplicated verb.
4. The Hittite type hassussara- ‘queen’ may allow us a new look at PIE *su̯ésor- ‘sister’.

The question of the IE family tree, especially with respect to the Indo-Hittite hypothesis, is a long and hotly debated issue of IE linguistics. In the paper, an overview of the most relevant features of Proto-Anatolian and Hittite will be given, which allow for a preliminary conclusion in this matter.

Recent research, most notably by Starke, Rieken, Rößle, and Widmer, has radically altered our picture of i-stems in Anatolian. Starke’s discovery of ‘i-mutation’ in the western Anatolian languages has refuted the notion that i-stems were rampantly productive in those languages. Rieken has shown that ‘i-mutation’ has nothing to do with either of the feminine ‘motion’ suffixes of Core PIE, while Widmer has made a convincing argument that the vṛkī suffix is present in Anatolian, but not in the function of marking the feminine. Rößle has established that oxtyone i-stem nouns were secondarily incorporated into the diphthongal stems in -āi-. No synthesis incorporating these new insights has yet been made, nor have certain still outstanding questions been addressed. I will seek to provide at least a provisional coherent overview of our current understanding of i-stems in Anatolian.