² This was our view at the time of the RAS Symposium. We now find a very satisfactory explanation based on the observation of Gelb, who pointed out (HH, III, 2) that the forms with the double strokes were written originally (and later by archaism) with the sign à (e for him) in place of the double stroke. Thus the double stroke is merely a cursive development of à written in ligature, and the -a-vocalization imparted by it to the original zi/a and i/ia, can be clearly understood.

³ Bossert, H. Th., “Ist die B-L Schrift im wesentlichen entziffert?”, Orientalia, XXIX, 1960, 423–442Google Scholar; ibid., XXX, 1961, 110–8. Cf. also “Zur Vokalisation des Luwischen”, ibid., XXX, 1961, 314–22.

⁴ cf. Özgüç, T., Altintepe II (Ankara, 1969)Google Scholar, Pl. LIII–LIV. [Cf. now Laroche, E., Anatolia, XV, 1971, 55 ff.Google Scholar; Klein, J., An. St., XXIV, 1974, 77 ff.]CrossRefGoogle Scholar

⁵ For the reading of the first two signs cf. Meriggi, Glossar, 238 and HH, No. 125.

⁶ See below, Appendix, p. 128, n. on 359.

⁷ cf. especially Mittelberger, H., “Bemerkungen zu Meriggis hieroglyphisch-hethitischem Glossar”, Die Sprache, IX, 1963, 68–107Google Scholar; “Zur Schreibung und Lautung des Hieroglyphenhethitisch”, ibid., X, 1964, 50–98.

⁸ It is also possible to show that internally and finally the ligature stands for ri+i and not for i+ra/i, or †a+ra (the traditional transcription).

⁹ For the reading †^Iī-ī-ā in Cekke see HHL, [47] n. 165; for the name †i-i-a which appears on a clay bulla from Korucutepe, cf. Güterbock, , JNES, XXXII, 1973, 143, Figs. 3–4.Google Scholar

¹⁰ In particular we have abandoned our reading of as a in favour of Neumann's ia.

¹¹ cf. Meriggi, P., Manuale di eteo geroglifico, (Roma, 1966, 1967), II, 87 ff.Google Scholar; see ibid., 95 f. for a list of the photographs on which Meriggi's edition is based.

¹² F. Steinherr, , “Die phönizisch-hethitischen Bilinguen vom Karatepe”, MSS, XXXII, 1974, 103–148.Google Scholar

¹³ We have only given the text of the Hu version; Ho, when relevant, is quoted in the commentary. Since our division into sentences differs considerably from that of Steinherr we have been obliged to alter the numeration; however, we have kept the same Arabic numbers for the individual words, though sometimes this is inconsistent with the word division which we have adopted.

¹⁴ For the form of this word see HHL, pp. [31, 34]; for our reading of the signs normally transliterated tí/tì see below, Addendum I.

¹⁵ The personal determinative which precedes caput is surprising but it is not easy to see how it should be explained.

¹⁶ For our reading of the name of Azatiwatas see HHL, p. [20 f.].

¹⁷ Sultanhan must now be read together with the new fragment published by DrEmre, K. in Anatolia, XV, 1971, 122 f. and Pl. XII.Google Scholar

¹⁸ It seems likely that another form of the same abstract REX-tahit- is attested in Aleppo 2, 2 where we read REX-^┌tá ?^┐-hi-tà, which may well be one of those -a forms mentioned below, p. 131 (g) (see the forthcoming edition of this inscription by Hawkins).

¹⁹ For the earlier proposals cf. Laroche, HH, 165; for Mittelberger's views see Die Sprache, IX, 1963, 82 n. 36.Google Scholar

²⁰ Sultanhan 9 has a form terra-tì-za followed by a small i sign which could belong to it or to the word that follows; a spelling terra-tì-i-za would be unique and it seems better not to give it too much importance, given the uncertainty of the reading. The Nom. plural terra-tì-i-zi (†ki-tì-a-i) found by Meriggi in Palanga, 2 is a reading far too dubious to serve as evidence.

²¹ That at some stage an equivalent of Hitt. pedan existed in the “Luwian group” is shown by Lycian. The new Greek-Aramaic-Lycian trilingual has a phrase pddẽhadẽ … pddẽnehmmis, κατστησε ἄρχοντας in which Laroche, , CRAIBL, 1974, 120Google Scholar, has recognized a preterite verb pdděhadẽ derived from the same root as pedan and comparable with Hittite pedaššaḫ-. Before the new evidence was available Laroche, , BSL, LXII, 1967, 61 f.Google Scholar had identified in the Xanthos stele a dative pddãt-i and a genitival adjective pddãt-ahi with the meaning “temenos”, which he had compared with Hittite pitta or rather with a supposed “Luwian” extension of pitta, pitant-. However, we may now wonder whether pddãt-i and pddãt-ahi should also be compared with pedan or with an extension ^*pedant-. (cf. also Carruba, O., Die satzeinleitende Partikeln in den idg. Sprachen Anatoliens, (Roma, 1969), 81 n. 77.Google Scholar)

²² For the word for “bad” see Hawkins, , An.St., XX, 1970, 88–89Google Scholar. It seems likely that the form quoted above is related to the forms of Karatepe 102, discussed below in 2. (a), (cf. also n. 23). (via) ha+ra/i-wa/i-ta-hi-tì-ti-i is a new reading in Kargamis A 15 b^**4: the word is the ablative of an abstract noun derived from harwa (n) t (a)- “road” (the exact form of this stem is still obscure: cf. Mittelberger, , Die Sprache, VIII, 1962, 185).Google Scholar

²³ See above n. 2. Not much can be made of Karatepe 375, malus-tí-sa-tara/i-ri+i, the ablative of an abstract noun built on the same adjective; even if the adjective had an -i- stem the vowel preceding the abstract suffix could be an a.

²⁴ In Sultanhan 4 C we expect the dative of mí-tí-sa and we find a form mi-tí-i. Does this speak against an -a value of tí? On the other hand the dative sing, of the obscure word á-tì-sa is written á-tì; could it be that the word has an -a- ending for the dative? For the dative sing. ending of the -a- stems, cf. HHL, p. [28].

²⁵ Meriggi's Glossar lists three forms whích would seem to speak in favour of an -i-value of tí (á-mi-tí, p. 26; †Á+s⁴-ī-tí-wá-tà-ā, p. 36; and †a-s³-tar-tí-a, p. 38); in all cases the correct reading is †ti ⁴ instead of tí.

²⁶ It is interesting to observe that the liquid signs seem to show some oscillations in their vocalism: cf. HH, No. 383: ra/i and HH, No. 445: lá/í/u.

²⁷ For references to ašša-, aššant- in Cun. Luwian see Laroche, , Dictionnaire de la langue louvite (Paris, 1959), 33Google Scholar with the corrections printed in Laroche, , RHA, LXXVI, 1965, 45Google Scholar; and the references mentioned there to Bossert, Meriggi, and Carruba.