Final -a

On pg. 148 I say that the fate of final -a  (from any potential PIE origin, i.e. *-a, *-h2e or *-h2) is unclear. I cite the case of ita, which, if equivalent to Ved. íti < *ith2, would show preservation of final -a.  However, I also note the possibility that it could come from *itā with Iambic Shortening.  

This hesitancy is inconsistent with the view I endorse firmly elsewhere that the athematic neuter nom./acc. pl. -a is from < *-h2 (e.g. p. 212). If the latter is true, then the outcome of final -a is definitively known. But this last fact is not absolutely certain.  In Sabellic it is clear that the thematic ending -ā < *-eh2 was generalized to athematic forms as Umb. tudero 'borders' VI a 15 (where tudero might be standing for /tuderof/, but the extension of the animate acc. pl. ending -f to the neuter presupposes a form tudero).  If this happened in Proto-Italic, independently, or by diffusion in the prehistory of Latin, and if there was a phonological shortening of final -ā, then the athematic neuter nom./acc. pl. would not be informative about the fate of final short -a.  But since trīgintā etc. seem to show that there was no shortening of final -ā, this scenario seems problematic.

In a recent article George Dunkel (2008) argues that final *-a became -e on the basis of the suffix -ne, as in superne, supposedly from *-na, which seems to match Hitt. ištar-na, Neo-Phrygian ενσταρνα. In addition the Umbrian form perne 'in front' is related to the adjective pernaio- and the most straightforward way to do this is to start from a pre-Umbrian *pernai.  This is an idea I have toyed with myself, but I don't feel confident that we can exclude a particle of the shape *-ne.

Dunkel also re-proposes an alternative etymology for nōn, i.e. by apocope from *nō-na, the *nō- of which can be compared with Hitt. natta 'not' < *no-th2(-oh1).  This is attractive since the traditional etymology from *ne-oinom requires a lot of special pleading.

Dunkel, Georg. 2008.  Chips from an Aptotologist's Worshop II. In Brigitte Huber, Marianne Volkart, and Paul Widmer (eds.), Chomolangma, Demawend und Kasbek. Festschrift für Roland Bielmeier zum 65. Geburtstag. Band II Demawend und Kasbek. Halle (Saale): International Institute for Tibetan and Buddhist Studies, 403–12.

Missing Reference

Matthew Scarborough points out to me that the reference on pg. 52 fn. 20 to  Mayrhofer 2005 does not correspond to anything in the bibliography.  The intended reference is:

Mayrhofer, Manfred. 2005. Die Fortsetzung der indogermanischen Laryngale im Indo-Iranischen. Vienna: Verlag der Österreichischen Akademie der Wissenschaften.

Thank you, Matthew!

Melchert Corrigenda XIII

The Hieroglyphic Luvian form that I quote as tamari 'builds' should be "quoted as as tammari* (AEDIFICARE+MI-ri+i). It is never spelled out, and by “Cop’s Law” the -mm- surely is geminate." This concludes the series of Melchert Corrigenda.  Thanks very much, Craig, for your help!

Corrigendum

On pg. 429 in discussing the paradigm of eō 'I go' I wrote "the 1st and 3rd singular have been thematized." Of course, this should read "the 1st singular and 3rd plural have been thematized." and in c.i on that page change 'expect' to 'except'.

Addendum

To the discussion of adverbs in -im with ablatival force add a reference to Goedegebuure 2007 who has demonstrated that Hieroglyphic Luvian had an ablatival suffix -in < -im in zin and apin from the pronominal stems za/i- 'this' and apa/i- 'that'.

Goedegebuure, P. 2007. The Hieroglyphic-Luwian Demonstrative Ablative-Instrumentals zin and apin. Studi Micenei ed Egeo-Anatolici 49:319–34.

Melchert Corrigenda XII

The explanation of the suffix -*sy- of the feminine oblique demonstratives from *-smy- needs to be reevaluated in light of the fact that a stem *syo- 'one' is required for Hittite and Tocharian.  See Pinault 2006.

Pinault, G-J. 2006.  Retour sur le numéral 'un' en Tokharien. Indogermanische Forschungen 110 71–97.

Melchert Corrigenda XI

On pg. 335 I wrote "Hittite evidence has been taken to suggest that the *so-/*to- pronoun originates in the combination of a sentence initial particle *su and a sentence-initial particle *to with an anaphoric pronoun stem *(h1)o-." But this is unlikely since, to quote a communication of Craig Melchert's, "šu has now been shown to correlate with past tense and ta with present-future. And the latter is with Rieken, MSS 59 (1999) 85-86, likely to be *toh1, an instrumental to the anaphoric stem. In any case, the Luvian “particle” *-sod attached to neuter nom.-acc. singulars shows that Anatolian had *so- as well as *to- as a

deictic/anaphoric stem. NB also dat. sg. enclitic *-soi. The peculiar suppletion in the *so-/to- paradigm is likely to be very ancient. Anatolian shows no trace of it because it almostentirely eliminated that entire pronoun in favor of *obho/i-."

Melchert Corrigenda X

On pg. 326 the Lydian form e~mi- (tilde over e) is wrongly cited as the accusative of the 1st person singular pronoun.  It is in fact the possessive adjective.  The real nominative and accusative is amu.

Melchert Corrigenda IX

On pg. 310–1 I endorse the claim of Jasanoff that  animate n-stems retained their n in the nominative singular  after ē in hysterokinetics but lost their final -n after ō, chiefly in amphikinetics.  However, the Hittite evidence does not support this view.  If Hittite MUNUS-anza 'woman' is to be read /kwanza/  reflecting a root noun *gwōns, that would suggest survival of n and recharacterization with s.  On the other hand if Hitt. aliyaš 'deer' reflects an n-stem nominative *h1olē+s. Of course, these facts do not settle the question of  the true distribution of n-ful and n-less nominatives. 

Addendum

On pg. 260 the Hittite form duwān (parā) 'long ago' should be added to Gk. δ(ϝ)ήν as a reflex of the e-grade*dweh2-m with Stang's Law. See Melchert 2008.

Melchert, H. Craig, 2008. Hittite duwān (parā). In Claire Bowern, Bethwyn Evans, and Luisa Miceli, eds., Morphology and Language History.  In Honour of Harold Koch. Philadelphia: John Benjamins, 201–9.

Corrigendum

On pg. 242 I cite the supposed Hittite genitive sg. itnaš in discussing the cognates of Lat. iter, itineris, but no such form is attested.  It is merely presumed that the word was an r-/n-stem.

Corrigendum

On pg. 235 in discussing the origin of the a-stem dative singular I write "PIE *-eh2iC ought to give -āi and *-eh2iV ought to give *-ai"  Of course, the conditioning environments are reversed.  In preconsonantal position the i would be syllabic and the h2 intervocalically would have been lost.  In prevocalic position the i would be nonsyllabic and the h2 would be lost in the coda with compensatory lengthening.

Addendum

For more examples of the neuter plural to animate singular—the locus ~ loca phenomenon—in Hittite alluded to on pg. 211 fn 24 see Melchert 2000:62–4.

Melchert, H. Craig. 2000. Tocharian Plurals in -nt- and Related Phenomena. Tocharian and Indo-European Studies 9.53-75.

Melchert Corrigenda VIII

The whole question of the endings of the ablative and instrumental has been reexamined in a far-reaching article by H. Craig Melchert and Norbert Oettinger, Ablativ und Instrumental im Hethitischen und Indogermanischen. Ein Beitrag zur relativen Chronologie.  The findings and suggestions of this article are extensive, but I should mention at least two.  

On pg. 207 fn. 7 I mention the idea that the PBS instr. pl. ending *-mīs might come from *-mins with an n imported from the accusative.  But this was never too likely.  Melchert and Oettinger instead suggest that *-mīs was modelled on an unattested but reconstructable *-bhīs < *-bhi-is. 

On pg. 212 I compare the Hittite instrumental suffix -it with the Vedic adverbial suffix -it. But Melchert and Oettinger argue that the underlying form of the suffix is -d as in the pronominal form ke-e-et ÍD-az /kēd hābats/ 'on this side of the river' etc. and that the i is epenthetic.  Hence there is no comparison between Hittite and Vedic. 

Corrigendum

Change the cross reference at pg. 205 fn. 2 from "2 b below" to "I B below"

Melchert Corrigenda VII

On pg. 203 I wrote "Hieroglyphic Luvian has -V-sa/i < -Vs(s)o as a thematic gen. sg. ending." but this is not quite accurate. HLuv. -Ca-si and Carian -ś reflect *-osyo, but HLuv. -CV-sa is ambiguous since it might continue *-o/eso or just *-os.

Melchert Corrigenda VI

On pg. 195 n. 12 add references to Hajnal 1994 and Rieken 2005 which have refuted the idea of  i-mutation as a reflex of  a PIE feminine formation.

Hajnal, Ivo 1994. Die lykischen a-Stämme: Zum Werdegang einer Nominalklasse. In  J.E. Rasmussen and B. Nielsen,  eds. In honorem Holger Pedersen. Wiesbaden:  Harrassowitz,135–71.

Rieken, Elisabeth, 2005. Neues zum Ursprung der anatolischen i-Mutation. Historische Sprachforschung 118:48–75.

Melchert Corrigenda V

On pg. 172 I give the proto-form of īnferus as *h1/3ndheros, but CLuv. ānnan ‘under, below’ argues definitively for *h1.

Corrigendum

On pg. 123 fn. 33 I wrote "one might unify the rules discussed in 4" instead of the correct "one might unify the rules discussed in D".

Clarification

On pg. 123 E I write "a short vowel following an r, l or n may be syncopated in an open syllable."  But the last example *sakros > sakrs > *sakerr > sacer  not conform to this formulation since the vowel following r in sakros is in a closed syllable. Of course, in sandhi the s would sometimes have been syllabified in the onset of the following word so r would potentially  be in an open syllable, but that seems a pis aller. Another possibility is that s was already "weak" in this position at the time of this the syncope. Finally, are we absolutely certain that this syncope would not have applied in closed syllables? What happens to the medial V in an original sequence *VCRVCCV?  I haven't come up with a certain case of this sequence yet.

Another point:  the sonorant consonant only develops to e/iR if it is "stranded"  in the nucleus after syncope.  If the sonorant became a coda by syncope no epenthetic vowel develops, e.g. *porkelelos > porcellus 'piglet'. 

Melchert Corrigenda IV

On pg. 123 and 169 I give the proto-form of vulnus as *welh2nos, but the root is more likely to be *welh3-.  Cf. LIV2 *welh3-.  Hittite walh- 'strike' could continue a medial *h3.

Corrigendum

On pg. 101, fn. 31 I wrote "the loss of w followed by a back vowel before monophthongization is suspect..." when I meant "the loss of w followed by a back vowel after monophthongization is suspect..."

Melchert Corrigenda III

On pg. 99 F replace Hitt. tuhhāi- 'groan' with Hitt. tuhhu(a)i- 'smoke' as the cognate of Lat. fūmus. The former may be onomatopoetic.

Corrigendum

On pg. 97, fn. 12 I wrote "*ēh3 and *ēh3 should also have given *ē, but there are no good examples." when I meant "*h3ē and *ēh3 should also have given *ē, but there are no good examples."

Clarification

On pg. 86 in discussing the out come of the PIE voiced aspirates in Germanic I write "In medial position these were realized as the voiced fricatives..."  In context the antecedent of 'these' is unclear and I should have written "In medial position the reflexes of the PIE voiced aspirates were realized as the voice fricatives..."

Melchert Corrigenda II

On pg. 54 (and again on pg 402 for another purpose) I cite the supposed Hittite form tarh- 'conquer' as an example of the preservation of a consonantal reflex of medial *h2 in Hittite, but Alwin Kloekhorst 2008:835–9 has shown that there is no such verb, but only a tarhu-/taruh- reflecting a u-present *terh2u-.

Further on pg. 54 I state that medial *h3 was loss without exception in Anatolian, but the question of the reflex of medial *h3 needs to be reconsidered. The meaning of CLuv. ta-tarh- as established by Kloekhorst is 'break' not 'conquer' and therefore might be compared with the family of Grk. τιτρώσκω ᾽I wound'. This suggests that *h3 may have had a consonantal reflex after a sononant.

Melchert Corrigenda I

Craig Melchert, the Diebold Professor of Indo-European at UCLA, has kindly forwarded a list of addenda and corrigenda which I will be posting beginning with this one.

On pg. 35 I quote Craig's famous 1987 article on the three-fold treatment of the voiceless velars in Luvian, but this needs modification.  Contrary to the claim of that article there is no good evidence for the affrication of the voiceless palatovelar before a back vowel. In particular, the supposed case of *-(i)k'o- > Luv. -(i)zza- ~ Lyc. -(i)s(e)- does not stand up. The Luvian suffix -(i)zza- probably goes back to *-tyeh2 and the Lycian suffix -(i)s(e)-, which makes abstracts and place names, continues *-sh2o-. Instead it now seems likely that there was a conditioned fronting of the voiceless palatovelar before front vowel (Luv. ziyari ‘lies’ (C) ~ Lyc. sijẽni), yod (wazi- ‘request’ (H) < *wek'-ye/o-), w (azu(wa)- ‘horse’ (H)< *ék'wo-) and syllabic sonorants (zurnid- ‘horn’ (H) < *k'r̥ng-id-, zanta 'down' < *k'm̥t-).  Thus Luvian and Lycian are like mirror images of Albanian with palatalization differential affecting the palatovelars. We still have a ternary reflex, just not an unconditioned one.

Update on Tagliavini's Diplomatic Service

In a previous post I speculated the Carlo Tagliavini's mission in Poland for the Italian foreign ministry in March of 1939 might have been a cover story for his involvement in preparations for the Italian invasion of Albania.  This is incorrect. In fact, the Italian foreign minister Galeazzo Ciano did visit Warsaw starting February 25, 1939 and this is presumably the mission which Tagliavini was involved in. Ciano hoped that war between Poland and Germany might still be avoided, but when he got to Warsaw he thought the Poles too resistant to giving in to German demands  (tutto rimane bloccato dall'antigermanismo qui radicato as he reported in a phone call to Mussolini ((Borejsza 1981:33)) and predicted war would not be long in coming. 

On the subject of Tagliavini during the run-up to World War II, I should mention that Tagliavini and, presumably through his influence, Ciano tried to get the German foreign ministry to request that the authorities allow the Austrian Albanologist of Jewish origins Norbert Jokl to flee to Albania where a position had been created for him.  But this intercession was unsuccessful and Jokl was to perish in Nazi hands.