Earliest Italic gen. sg. in -i

On. pg. 234 fn. 10 I mention that the form MORAI from a 4th century inscription from Signia or the Ager Signinus (the place is called Signia not Signum as I write there and is modern Segni about 60 km southeast of Rome) is the earliest example of gen. sg. -ī in any stem class. However, this is no longer true. In Zeitschrift für Papyrologie und Epigraphik 168, 2009, pp. 273–7 Maria Cristina Biella ("Una nuova iscrizione falisca di VII a.C.: Un sostantivo con tema in -o e genitivo in -i") has published a 7th century Calix from Falerii with the inscription TITI. This form is most probably a genitive of the common praenomen Titos and  might be taken as more evidence that the original locus of the -ī genitive was in names.

Corrigenda varia

pg. 201, ln. 6 Correct the Serbo-Croatian vocative of 'woman' from žȅna to žȅno.


pg. 211 The content of note (l) about the origin of the Lith. instr. sg. -mì should be moved to note (j)

pg. 249  In the chart of PIE endings for the u-stems beside the lengthened grade endingless locative in
*-ēu (or *-ēu̯ as I probably should have written it) add the ending-ful locative *-eu̯-i.

In the index on pg. 633 under the heading pius ~ purus rule add 142 and change 198 to 191.

Thanks to Thomas Olander and Ben Forston for pointing out these mistakes.

Corrigendum

Ben Fortson points out that on p. 446 n. 81 I say that Gerhard Meiser in Veni Vidi Vici 2003:57 compares the Latin passive infinitive in -ier with the Vedic gerundive in -ya, but this is not an accurate presentation of Professor Meiser's views.  In fact, he prefers the idea that a form like *amā-đyēr, with the  cognate of Umbrian passive infinitive suffix -fi, Osc. -fír, was transformed into *amā-zyēr (> amārier) after the active *amā-zi (> amāre). This idea is certainly worthy of serious consideration. Meiser did uphold the comparison with the Vedic gerundive in 1998:225.

Corrigendum

On pg. 283 fn. 98 I mention the adverb lūdicrē which occurs in Ennius, but the Ennius passage preserved by Nonius p. 195.1 L reads: pars ludicre saxa iactant, inter se licitantur which must be divided as Skutsch 69-70 and others do after saxa:

 pars ludicre saxa 

 iactant, inter se licitantur

This means, as Brent Vine pointed out to me, that the final e of ludicre is short and probably is to be taken as the neuter of an adjective ludicris* which Priscian Keil 2.350 (ab hoc ludicri) attests. The OLD gets this right s.v. ludicre.

Addendum

To the very brief discussion of aphaerisized forms of esse on pg. 426 n. 4 add the reference Pezzini 2011, which gives a very nice collection of evidence from the manuscripts and inscriptions. 

Pezzini, Giuseppe. 2011. "Contraction of est in Latin." Transactions of the Philological Society 109:327–43.

Second Printing!

The second corrected printing is now available.  If your library doesn't have it yet, get them to order it.

Another ghost word

On pg. 241 I cite the Hittite form itar, itnaš as a cognate of Lat. iter, itineris, but this word does not exist. Its one alleged occurrence (KUB 41.8 i 20) is now read as DUMU-tar 'offspring'.  See Jared Miller 2008. "Ein Ritual zur Reinigung eines Hauswesens durch eine Beschwörung an die Unterirdischen (CTH 446)" in B. Janowski and G. Wilhelm (eds.) Texte aus der Umwelt des Alten Testaments, N.F. vol. 4, p. 209, fn. 97. The Tocharian cognates remain safe...for now.