Hostname: page-component-586b7cd67f-t7fkt Total loading time: 0 Render date: 2024-11-23T20:38:44.526Z Has data issue: false hasContentIssue false
  • English
  • Français

The Ablaut of Present and Preterite in Hittite Radical Verbs

Published online by Cambridge University Press:  23 December 2013

Gillian R. Hart
Get access Rights & Permissions [Opens in a new window]

Extract

The extent to which Hittite has preserved Indo-European ablaut patterns has always been a matter for debate. Variations in the vowel of the root occur in both the -mi and the -ḫi conjugations, although roots of the shape TeRT- have eliminated vocalic variation altogether, and certain other verbs, such as pai- “go” have similarly only one form of the root. In those roots which end in a single consonant and show variation in the root vowel several types may be distinguished. In the -mi conjugation the type represented by 3 sg. ku-en-zi, 3 pl. ku-na-an-zi or 3 sg. ku-e-ir-zi, 3 pl. ku-ra-an-zi has always been recognized as reflecting inherited e/zero alternation, as seen also in Sanskrit 3 sg. hánti, 3 pl. ghnánti. This type presents no difficulties from the comparative point of view except for the appearance of the e-grade of the root in some persons where zero grade might have been expected, as in the preterite plural and 2 pl. imperative, which is identical in form with the 2 pl. indicative of the preterite.

Type
Research Article
Information
Anatolian Studies , Volume 30 , December 1980 , pp. 51 - 61
Copyright
Copyright © The British Institute at Ankara 1980

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Where T represents a consonant (stop, laryngeal or *s), and R one of the semivowels *r, *l, *m, *n.

2 For example ku-e-u-en KUB XXIII 21 obv. 15; ku-en-te-en KUB XXIII 68 obv. 26′; ku-e-ni-ir KBo III 60 iii 9. For descriptions of the ablaut alternations prevailing in the -mi conjugation cf. Kronasser, H., Etymologie der hethitischen Sprache, Wiesbaden 1962–1966, p. 21Google Scholar, with references to earlier literature, and Kammenhuber, A., Altkleinasiatische Sprachen = Handbuch der Orientalistik: Erste Abteilung, Zweiter Band, Erster und Zweiter Abschnitt, Lieferung 2, Leiden/Köln 1969, pp 224–228Google Scholar.

3 For the various environments in which vowel loss did or did not occur at the time of the IE quantitative ablaut changes see Kuryłowicz, J., Indogermanische Grammatik II, Heidelberg 1968, pp 208–211Google Scholar.

4 Although ú-e-ḫa-an-zi is also attested, the antiquity of wa-ḫa-an-zi within Hittite is guaranteed by its appearance in texts with old ductus, for example in KBo XX 19, 13Google Scholar; XX 20 obv. 11; KBo XXV 31 ii 14′. Especially noteworthy is the form wa-ḫa-a-an-zi in KBo XXV 42Google Scholar left 15, which provides a further example of plene-spelling in the termination of the 3 pl. present of an ablauting verb of the -mi conjugation; I have discussed other examples of the kind, and suggested that the plene-spelling reflects the position of the word-accent, in BSOAS XLIII (1980:1), pp. 1–17Google Scholar. Despite the antiquity of these forms with -a-vocalism, it is possible that the active ablauting verb is itself an innovation; it must be so if it is to be derived from IE *weiH 2, as proposed by Eichner, H., “Die Etymologie von heth. meḫur”, MSS XXXI, (1973), p. 76Google Scholar. In favour of Eichner's etymology is the fact that this verb has a mediopassive with invariable e-grade of the root; for the forms see Neu, E., Die Interpretation der hethitischen mediopassiven Verbalformen, StBoT 5, Wiesbaden 1968, pp 195–199Google Scholar. The active, which like the mediopassive is intransitive, may well be a secondary development based directly on the mediopassive, with vocalic alternation and shifting accent modelled on the existing type with e/a alternation. A similar case of secondary ablaut alternation is likely in wes- “wear”, which has been made the subject of a special study by Eichner, H., “Hethitisch (Gewänder) tragen, anziehen, bekleiden”, MSS XXVII, (1970), 5–44, especially p. 27 and p. 41Google Scholar, note 50.

5 Kuryłowicz, J., Indogermanische Grammatik II, p. 255Google Scholar, linked this development with the fact that in roots of the form TeRT- the zero grade would have had the same outcome in Hittite as the o-grade (TaRT- from both TRT- and ToRT-), so that (T)at- came to be an appropriate representation for the zero grade of (T)eT- roots. Kuryłowicz further held that the vowel which appeared in the zero grade of (T)eT- roots was originally e, and that this vocalism was retained in Hittite in forms of the 1 and 2 pl. like eppueni, epteni. The chronology of the Hittite forms, however, is against such a view.

6 It was felt as a difficulty by H. Kronasser, who thought that it involved positing an unlikely amount of e/b or ē/ə ablaut; this, together with the problems raised by the distribution of the variants, made him inclined to prefer a theory of vowel assimilation. (Etymologie der hethitischen Sprache p. 21, 24).

7 Its chief advocate was Petersen, Walter, whose article “Vowel assimilation in Hittite”, JAOS LIV (1934), 161–168Google Scholar contained a criticism of Sturtevant's explanation in terms of ablaut, and proposed an alternative to it. The idea also received some support from Pedersen, H., Hittitisch und die anderen indoeuropäischen Sprachen, København 1938, pp 128, 167Google Scholar, and from H. Kronasser (cf. note 6 above).

8 Language XI, (1935), pp 175–184Google Scholar.

9 A very full collection of forms has now been published for this verb by Karamenhuber, A., Materialen zu einem hethitischen Thesaurus, Heidelberg 1973Google Scholar–; paradigms are given in Lieferung 4 (1975), pp. 81–82Google Scholar.

10 a-ku-e-ni StBoT 8 ii 13, iii 15, iv 6; KUB XXXVI 110 rev. 6; Bo 4194. Also KBo III 29 i 19 (an Old Hittite text in a later copy), and KUB XLVI 65Google Scholar obv. 6.

11 e-ku-e-ni 412/b+ obv. (35b), 37b. A form a-ku-wa-ni quoted by Otten, H. and Souček, V. from Bo 5709 obv. 10 (Ein althethitisches Ritual für das Königspaar = StBoT 8, Wiesbaden 1969, p. 78Google Scholar) looks like an innovation parallel to e-ku-wa-ni but without the change in root vocalism. Cf. a-du-wa-ni from the same line of the same text.

12 Carruba, O., Das Beschwörungsritual für die Göttin Wišurijanza = StBoT 2, Wiesbaden 1966, p. 46Google Scholar.

13 a-tu-e-ni StBoT 8 iii 15, iv 6; a-du-e-ni ii 13; KUB XXXVI 110 rev. a-tu-e-ni KBo III 29 i 19. On a-du-wa-ni see note 11.

14 As for example in KUB XXXI 42 iii 19; KUB VII 5 ii 4. In the Hukkanas-treaty the same variants occur at KBo V 3 iv 12 pa-a-i-u-a-ni and 13 pa-a-i-wa-a[-ni]; the language of the treaty has been recently discussed by Neu, E., “Zum sprachlichen Alter des Hukkana-Vertrages”, KZ XCIII, (1979), pp. 64–84Google Scholar. Neu concludes that the language of the treaty can be classified as Middle Hittite, but ms. A, of which KBo V 3 forms part, is a thirteenth-century copy, although usually a careful one. Some texts in the old ductus, on the other hand, have the simpler spelling pa-i-wa-ni; so regularly in StBoT 8, and also in KBo XXII 2 i 15; KUB XLIII 33 obv. 4, and KBo XVII 25 obv. 6′ “in ziemlich alter Schrift”.

15 For the distinction between these particles see Rosenkranz, B., “Die hethitischen Schreibungen für ‘und’“, Symbolae Biblicae et Mesopotamicae Francisco Mario Theodoro Böhl dedicatae, Leiden 1973, pp. 320–326Google Scholar, and Houwink ten Cate, P. H. J., “The Particle -a and its Usage with Respect to the Personal Pronoun”, Festschrift Heinrich Otten, Wiesbaden 1973, pp. 119–139Google Scholar.

16 StBoT 8, pp. 77–78Google Scholar.

17 For example a-ú-me-en KBo XXII 129 obv. 4, 10, KUB XVIII 5 i (23), 33, (36), 46; KUB IX 34 iii 39, 40, 41; a-uš-ten KUB XV 34 ii 12. Further examples from unpublished texts are given by Otten, H., Eine althethitische Erzählung um die Stadt Zalpa = StBoT 17, Wiesbaden 1973, p. 30Google Scholar.

18 A late example is da-a-it-te-ni KUB XIII 6 ii 5.

19 Laroche, E., “La stèle trilingue du Létôon; Deuxième partie, l'inscription lycienne”. Fouilles de Xanthos, Tome VI, Paris 1979, p. 106Google Scholar.

20 For a synopsis of the forms from this verb see von Schuler, E., Die Kaškäer, Ein Beitrag zur Ethnographie des alten Kleinasien, Berlin 1965, p. 126CrossRefGoogle Scholar.

21 For example KBo III 17 obv. 7; KBo XXIV 24 iii 22′; KBo XXV 68 i 12′. as well as the more ambiguous spelling pí-i-ir, KBo III 3 3 ii 8; KBo III 34 ii 18; KBo III 55 obv. 10; KBo VIII 4 2 rev. 10.

22 pí-ú-e-ni also at KUB II 2 ii 50; KBo XVI 50 obv. 6; pí-i-ú-e-ni KBo XXI 21 rev. 15′.

23 Jasanoff, J. H., “The Position of the -ḫi Conjugation”. Hethitisch und Indogermanisch: Vergleichende Studien zur historischen Grammatik und zur dialektgeographischen Stellung der indogermanischen Sprachgruppe Altkleinasiens. Herausgegeben von Erich Neu und Wolfgang Meid. Innsbruck 1979, pp. 79–90Google Scholar.

24 As for example by Risch, E., “Zur Entstehung des hethitischen Verbalparadigmas”, Flexion und Wortbildung: Akten der V. Fachtagung der Indogermanischen Gesellschaft. Regensburg, 9.–14. September 1973. Herausgegeben von Helmut Rix. Wiesbaden 1975, pp. 247–258, esp. p. 256(e)Google Scholar. The same view is taken by Lindeman, F. O., “L'apophonie radicale au present-imparfait actif des verbes athématiques en indo-européen”, BSL LXXI, (1976), pp. 113–121Google Scholar.

25 See in particular the contributions to Hethitisch und Indogermanisch by Cowgill, W. “Anatolian -ḫi conjugation and Indo-European Perfect: Instalment II”, pp. 9–40Google Scholar and Meid, W. “Der Archaismus des Hethitischen”, pp. 159–176Google Scholar.

26 The antiquity of the full grade root in the aorist has been maintained by Hoffmann, K., “zum Optativ des Indogermanischen Wurzelaorists”, Pratidānam: Indian, Iranian and Indo-European Studies Presented to F. B. J. Kuiper on his Sixieth Birthday, The Hague 1968, pp. 3–8Google Scholar; this view has been contested by Francis, E. D., “Greek ἔβλην”, Glotta LII, (1974), pp. 11–30Google Scholar, because of the difficulty in explaining the Homeric dual βάτην “they two went” (beside βήτην) as an innovation. The traditional view is also supported by Kuryłowicz, J., Indogermanische Grammatik II, pp. 81–82Google Scholar, who regards the Greek and Indo-Iranian phenomena as resulting from independent innovations. Kuryłowicz likewise regards the Vedic imperatives of the 2 pl. with full grade of the root as innovations, although Watkins, C.. Indogermanische Grammatik III/1, pp. 32–33Google Scholar is inclined to regard the imperatives (though not the aorists) as archaic survivals. For an attempt to explain the Vedic 2 pl. imperatives as having acquired full grade of the root under the influence of the optative see the article by F. O. Lindeman referred to in note 24. Eichner, H., “Die Vorgeschichte des hethitischen Verbalsystems”, Flexion und Wortbildung p. 82Google Scholar follows K. Hoffmann in regarding the full grade radical vocalism of the Vedic and Greek aorists as archaic, and finds support for this in the Hittite situation, where he believes that the ablaut inherited in aorist stems was extended to the past tenses of old root presents like kuen- “kill”.

27 For these see the article of Eichner cited in note 26 above, especially p. 82.

28 Indogermanische Grammatik III/1, pp. 24–25Google Scholar.

29 Cf. the discussion by Watkins, , Indogermanische Grammatik III/1, p. 35Google Scholar.