¹ Die Sprache der Hethiter, Boghazköi-Studien 1–2 (Leipzig, 1916–7). Only the comparisons based on a phonetic development of IE ĝh- to Hitt. ḫ are put forward unreservedly, e.g. 12 fn. 2: Hitt. ḫalki- ‘grain’ < IE *ĝhel- ‘green’ (Gk. χλóη ‘young shoot’, Phryg. ςέλκɩα Vegetables', OCS zlakъ 'grass'); 13 fn. 2: Hitt. ḫar- (claimed to mean 'seize, grasp') < IE *ĝher- 'seize' (Skt. hárati ‘takes, seizes’, Gk. χεíρ 'hand'). When connecting Hitt. ḫalzā(i)-(with the alleged meaning 'shout'; but cf. Ferdinand Sommer, Hethitisches 1.10–1, Boghazköi-Studien 4 [Leipzig, 1920]) with Czech hlásati ‘shout’ (op.cit. 40 fn. 3) and deriving Hitt. ḫandā(i)- ‘establish’ from IE *ghend- ‘grasp’ (ibid. 11 fn. 1), Hrozný points out the divergent treatment of the IE consonant in medial position and stresses the wide difference in meaning.

² Caractère indo-européen de la langue hittite 144–9 (Christiania, 1919); discussed by Walter Couvreur, De hettitische Ḫ, een bijdrage tot de studie van het indo-europeesche vocalisme 61–3 (Leuven, 1937; hereafter abbreviated Hett. Ḫ).

Most of Marstrander's evidence is borrowed directly from Hrozný, e.g. the derivation of Hitt. ḫarrant- (= Akkad. ṣapru) from ḫar- ‘seize’ (: Skt. haraka- 'taker, rogue'), and the connection of Hitt. ḫaršalant- 'angry' with Skt. kárṣate, kŕṣyati ‘is excited’, Lat. horreō (cf. Sprache der Hethiter 1.86 fn. 4–5).

³ ZDMG NF 1.159 (1922); OLZ 26.49 (1923). Mention may also be made of Albrecht Goetze's interpretation of Hitt. weḫiški/a- ‘versari, vigilare’ as a cognate of Goth. wakan, Madduwattaš 107 fn. 3, Hethitische Texte in Umschrift 3 (Leipzig, 1928)—rightly rejected by Couvreur, Hett. Ḫ 66; as well as the unacceptable views of Kellogg referred to by Walter Petersen, Lg. 10.307 fn. 3 (1934), and Couvreur, op.cit. 68.

⁴ Groupement des dialectes indo-européens 42 (Copenhagen, 1925); cf. also Litteris 5.159 (1928). Giuliano Bonfante still holds this view in CP 39.52 fn. 4 (1944).

⁵ In connection with Jerzy Kuryłowicz, Symbolae grammaticae in honorem Ioannis Rozwadowski 1.101 (Cracow, 1927; abbreviated Symb. Gramm.); Litteris 5.159 (1928), but with some hesitation. Cf. also AOr. 5.177 ff. (1933), discussed by Couvreur, Hett. Ḫ 64 f.

⁶ Op.cit. 101; cf. Edgar H. Sturtevant, Lg. 6.151 (1930); Émile Benveniste, Origines de la formation des noms en indo-européen 155 fn. 1 (Paris, 1935; abbreviated Origines); Couvreur, Hett. Ḫ 120 ff.; etc.

⁷ Hittitisch und die anderen indoeuropäischen Sprachen 138 (Copenhagen, 1938; abbreviated Hitt. u. IE); rejected by Sommer, Hethiter und Hethitisch 91 (Stuttgart, 1947; abbreviated Heth. u. Heth.). A satisfactory IE etymology of Hitt. ḫatrāmi has now been supplied by Michel Lejeune, BSL 46.43–7 (1950), who compares the Hittite verb to Venetic atraest ‘sculpsit’ (or ‘mandauit‘?).

⁸ Pedersen, op.cit. 176, 179 n.

⁹ Ibid. 177 fn. 1.

¹⁰ Cf., for parallels, Pedersen, Hitt. u. IE 178.

¹¹ Ibid. 201.

¹² Untersuchungen über die Bedeutung des Hethitischen für die Laryngaltheorie 25 (Copenhagen, 1941; abbreviated Laryng.).

¹³ Op.cit. 26–7.

¹⁴ TPhS 86–7 (1946).

¹⁵ Lykisch und Hittitisch 28 f. (Copenhagen, 1945; abbreviated Lyk. u. Hitt.).

¹⁶ Laryngeal before sonant 55–8 (Copenhagen, 1948; abbreviated Laryng. bef. Son.).

¹⁷ Hammerich, op.cit. 56.

¹⁸ Ibid. 57.

¹⁹ Az indoeurópai ínyhangok a hetitában 398–400, Melich-Emlékkönyv (Budapest, 1942).

²⁰ According to Ehelolf, OLZ 36.6 (1933).

²¹ Which Hendriksen's etymology does not necessarily imply, however, as he himself (Laryng. 59) posits *g^wiH-dhor.

²² Lg. 16.85 (1940).

²³ BSL 33.137 (1932).

²⁴ Cf. Pedersen, Hitt. u. IE 130.

²⁵ In his Beiträge zur hethitischen Grammatik 15–6 (Budapest, 1942), however, Szemerényi restates the comparison with Gk. ὄλλυμɩ ‘destroy’ put forward by Couvreur, Hett. Ḫ 143 f. He considers Hitt. -ul- to be the phonetic reflex of IE *ḷ and accordingly derives Hitt. ḫulla- from IE *H ₃ḷ-neH ₂-.

²⁶ In Goetze, The Hittite ritual of Tunnawi 70 (New Haven, 1938; abbreviated Tunnawi).

²⁷ Cf. Couvreur's similar objection (Hett. Ḫ 59) to the same connection in Hrozný's Sprache der Hethiter (fn. 1 above).

²⁸ Szemerényi, Melich-Emlékkönyv 387 (Budapest, 1942).

²⁹ Cf. also the adverb and postposition ḫanda ‘actually; to be sure’ and ‘according to’, in fact the accusative plural neuter of a participle *ḫanza; Szemerényi, Anal. Orient. 1.13 (1942).

³⁰ Lyk. u. Hitt. 27 f. (1945).

³¹ Lg. 24.315 f. (1948).

³² Sturtevant, Lg. 17.184 (1941); The Indo-Hittite laryngeals §36i (Baltimore, 1942; abbreviated IHL); Lg. 19.308 (1943); A comparative grammar of the Hittite language 1.30, 48 (rev. ed., New Haven, 1951; abbreviated HG²). This etymology was first proposed by Kuryłowicz, Symb. Gramm. 1.102 (1927). A divergent view is expressed by Couvreur, Hett. Ḫ 218–20.

³³ In addition to Pedersen's reserves (Lyk. u. Hitt. 28), see Piero Meriggi's interpretation as ‘gens’, Germanen und Indogermanen, Festschrift für H. Hirt 2.260 (Heidelberg, 1936); cf. Paul Kretschmer, Glotta 27.261 (1939).

³⁴ Pedersen, Lyk. u. Hitt. 14, 25 f., 43.

³⁵ Sturtevant, HG¹ 72 (1933); Couvreur, Hett. Ḫ 50–1 (1937).

³⁶ Emmanuel Laroche, Recherches sur les noms des dieux hittites 28, 42 (Paris, 1947; abbreviated Dieux hittites).

³⁷ J. Friedrich, Hethitisches Elementarbuch 1.8 (Heidelberg, 1940); Albert Cuny, RHA 6.79 (1942–3).

³⁸ A. Goetze, Die Annalen des Muršiliš 204 fn. 1, Hethitische Texte in Umschrift 6 (Leipzig, 1933); Knut Bergsland, RHA 4.272 (1938).

³⁹ F. Sommer, Aḫḫijavā-Frage und Sprachwissenschaft 73–93 (Munich, 1934); Heth. u. Heth. 24 (1947).

⁴⁰ JAOS 59.186–9 (1939).

⁴¹ Sturtevant, Lg. 12.184–5 (1936); IHL §46c, §71c. Contrary to the views still advocated in HG² 50 (1951), I cannot bring myself to believe that Hitt. -eḫ- reflects original *-eγ- and corresponds to *ē in the cognate IE languages. My reasons for disagreeing with Sturtevant on this point are set out in my article Zum heutigen Stand der Laryngaltheorie, RBPhH 30.459–63 (1952); I have given a different explanation of the facts in Handelingen der Zuidnederlandse Maatschappij voor Taal- en Letterkunde en Geschiedenis 4.69–70 (Oudenaarde, 1950). I connect Hitt. šeḫur: OIcel. saurr with Illyr. Sava (river name), OE sēaw ‘juice, moisture’, Icel. söggr ‘dank, wet’, etc.

⁴² Sturtevant, IHL 47 fn. 38. Hitt. mi-ya-ḫu-wa-an-ta-ḫu-ut ‘grow old’, is related to meḫur as Av. zvran- ‘time, point of time’ (cf. Benveniste, BSL 34.22–4 [1933]) is related to zaurvan- ‘old age, weakness of age’ (: Gk. γέρων 'old man'); for this reason I think the Hittite e stands in this word for an original i-diphthong, and cannot accept Sturtevant's derivation from the root *meγ- 'measure' (IHL §46b, HG² 50).

⁴³ Sturtevant, Lg. 7.120 (1931), IHL §36b, HG² 46–7.

⁴⁴ Pedersen, Muršiliš Sprachlähmung 61 (Copenhagen, 1934; abbreviated MS); Couvreur, Hett. Ḫ 183–4; Sturtevant, IHL §36d, HG² 47.

⁴⁵ Mélanges linguistiques offerts à M. Holger Pedersen 492 fn. 3 (Aarhus, 1937); cf. Sturtevant, Lg. 14.76 fn. 5 (1938).

⁴⁶ Ernst Risch, Wortbildung der homerischen Sprache 163 (Berlin, 1937).

⁴⁷ Cf. Carl Darling Buck, A dictionary of selected synonyms in the principal IE languages 212–3 (Chicago, 1949), to which may be added OIr. mullach ‘summit, head’ (Pedersen, Vergleichende Grammatik der keltischen Sprachen 2.55 [Göttingen, 1913]); A. J. J. De Witte, De betekeniswereld van het lichaam: Taalpsychologische, taalvergelijkende studie 86–9 (Utrecht–Antwerp, 1948).

⁴⁸ This possibility is excluded by Couvreur's assumption that IE *ó is reflected by o in both Tocharian dialects; see Hoofdzaken van de tochaarse klank- en vormleer 10 f., 80 (Leuven, 1947). Cf. George S. Lane, Lg. 24.311 (1948). Duchesne-Guillemin, BSL 41.173 (1941), gives a different explanation of the Tocharian word.

⁴⁹ Sturtevant, in Goetze, Tunnawi 70 (1938), translating the sequence ḫar-aš-zi te-ri-ip-zi ‘he tills (and) plows’, whereas in HG² 15 (1951) ḫaršzi is interpreted as ‘he harrows’, probably on account of its etymological connection with Hitt. ḫarra- ‘break, pulverize’ (cf. IHL §37f).

⁵⁰ Laryng. bef. Son. 25–6, 56, 69.

⁵¹ E. Raucq, Bijdrage tot de studie van de morphologie van het IE verbum 58 fn. 1 (Bruges, 1951). With Raucq, I am inclined to posit an IE noun *d_emәs, which Vittore Pisani, Crestomazia indeuropea 113 (Turin, 1947), also identifies in Ved. dāsáḥ ‘slave’ (cf. my review of Raucq's work, RBPhH 29.1204 (1951)).

⁵² Cf. Gordon M. Messing, Glotta 31.249–50 (1951).

⁵³ Sturtevant, HG² 65.

⁵⁴ Lyk. u. Hitt. 53–4; cp. F. J. Tritsch, AOr. 18, 1.516–7 (1950).

⁵⁵ A. Meillet and J. Vendryes, Traité de grammaire comparée des langues classiques² 289 (Paris, 1948).

⁵⁶ F. B. J. Kuiper, Acta orientalia 12.190 ff., esp. 210 ff. (1934); Die indogermanischen Nasalpräsentia 36–40, 63 fn. 2 (Amsterdam, 1937); Pedersen, Hitt. u. IE 95.

⁵⁷ Sturtevant, in Goetze, Tunnawi 70 (1938).

⁵⁸ Cf. Jost Trier, Btr. 67.110–22 (1944).

⁵⁹ Cf. RBPhH 26.558 fn. 13 (1948). In spite of Walde-Hofmann 1.29, the Latin word is probably not borrowed (note the initial h- in halicarius [Lucilius 15]: Gk. ἄλɩξ).

⁶⁰ As O. Szemerényi kindly points out to me.

⁶¹ Laroche, Dieux hittites 73 (1947).

⁶² Sommer, Heth. u. Heth. 95 (1947). In this case, Lat. alica should be related to the root of Maced. (cf. A. Ernout and A. Meillet, Dict. étym. de la langue latine³ 1.32 [Paris, 1951]), rather than to Lith. alùs, OIcel. ǫl, etc. ‘beer’ (cf. V. Pisani, Testi latini arcaici e volgari 94 fn. [Turin, 1950]), on account of the parallel connection of Gk. ἄλφɩ ‘peeled barley’, Alb. elp ‘barley’ to IE *albho- ‘white’; cf. Franz Specht, Der Ursprung der indogermanischen Deklination 68, 114–5 (Göttingen, 1944).

⁶³ Cf. RBPhH 29.160 fn. 3 (1951); Mélanges Henri Grégoire 2.566 fn. 2 (Brussels, 1950).

⁶⁴ Couvreur, Hett. Ḫ 59 (1937).

⁶⁵ Vocabulaire étymologique de la langue hittite 20 (Limoges, 1942).

⁶⁶ Laroche, Dieux hittites 21 (1947). Szemerényi suggests a borrowed derivative *aluka ‘goer’ from Akkad. alāku ‘go, come’ as the source of the ‘Nesite’ word. To this, Kupper objects that the only forms of Akkad. alāku to be attested are āliku ‘going’, always used with a genitival phrase (e.g. ālik maḫri 'going first'), and allāku, sometimes meaning 'messenger', for which the usual words are rakbu and mar šipri.

⁶⁷ Symb. Gramm. 1.102 (1927).

⁶⁸ Lg. 6.152 (1930).

⁶⁹ MS 74 (1934).

⁷⁰ Hitt. u. IE 201, cf. 157, 177–8 (1938).

⁷¹ Vergleichendes Wörterbuch der gotischen Sprache 212 (Leyden, 1939; abbreviated VWGS).

⁷² Hett. Ḫ 211 (1937).

⁷³ Lg. 14.109–10 (1938).

⁷⁴ Lg. 16.82 (1940); cf. HG² 50.

⁷⁵ IHL §46d (1942).

⁷⁶ RHA 4.270–4 (1938).

⁷⁷ I have set forth my views on the subject in the two articles referred to above in fn. 41.

⁷⁸ Cf. Sturtevant, HG¹ 140 (1933). Couvreur, Hett. Ḫ 221–2 (1937), rejected this view on account of the assumed meaning ‘scatter, empty’, but most scholars now agree in translating it as ‘pour’.

⁷⁹ HG² 51, with fn. 87 (1951).

⁸⁰ Lg. 15.150–1 (1939). As the root *sed- also appears in the zero grade with a different prefix in Hitt. pár-aš-du-uš ‘young shoots’ (KUB 17.10.1.16), Sturtevant feels entitled to disregard Couvreur's objection to his interpretation on account of Ehelolf's translation ‘leaves’ (Hett. Ḫ 147).

⁸¹ Cf. Couvreur, Hett. Ḫ 155.

⁸² Lg. 15.152 (1939).

⁸³ Henry G. Liddell and Robert Scott, A Greek-English lexicon⁹ 1.40 (Oxford, 1925).

⁸⁴ Émile Boisacq, Dictionnaire étymologique de la langue grecque 26 (Heidelberg, 1916); J. B. Hofmann, Etymologisches Wörterbuch des Griechischen 7 (Munich, 1950).

⁸⁵ Boisacq, ibid.; but cf. Feist, VWGS 29.

⁸⁶ Pending further evidence of the correspondence of Lyc. q to Hitt. ḫ, we must leave open the question whether Pedersen's connection of Lyc. qastti qanuweti ‘avenges, punishes’ with Hitt. ḫanna- fits this etymology.

⁸⁷ Laryng. 57, 59 (1941); but cf. Cuny, RHA 6.76 (1942–3).

⁸⁸ Cf. Hitt. nanšan (KUB 23.11.3.11) instead of naššan; Friedrich, Heth. Elementarb. 1.9 (1940).

⁸⁹ Cf. Fritz Mezger, Lg. 15.189 (1939).

⁹⁰ Cf. Tritsch's argument, AOr. 18, 1.516–7 (1950), against Pedersen's interpretation of the word, Lyk. u. Hitt. 53–4 (1945).

⁹¹ MS 57 (1934).

⁹² RBPhH 28.1427 (1950); cf. my forthcoming article, L'étymologie du terme germanique *ansuz 'dieu souverain', Ētudes germaniques (1952).

⁹³ Lg. 16.85 (1940); IHL §47i (1942); Lg. 20.208 (1944); HG² 50 f., 71 (1951).

⁹⁴ HG¹ 150 (1933); cf. Alfonsina Braun, Il lessico ittita nei suoi reflessi indoeuropei 394–5 (Venice, 1936).

⁹⁵ Sturtevant, HG² 72 (1951).

⁹⁶ Antiquité classique 12.108 (1943).

⁹⁷ As Szemerényi has kindly suggested to me.

⁹⁸ Couvreur, Hett. Ḫ 119 (1937). William M. Austin, Lg. 17.90 (1941), erroneously ascribes this etymology to Benveniste, Origines 155. Sturtevant, who abandoned his former analysis as *ho-we-ēi (HG¹ 94) in Lg. 15.154 fn. 31 (1939), seems to favor it again in HG² 128 (1951).

⁹⁹ Cf. fn. 25.

¹⁰⁰ Lg. 18.191 (1942); IHL §361.

¹⁰¹ IHL §66c.

¹⁰² Cf. my additional note, Mélanges Grégoire 2.568–9 (1950). At any rate, the assumption of a dissimilative loss of Hitt. ḫ- in ḫwalḫḫ- (Szemerényi, Melich-Emlékkönyv 400 [1942]) is no more convincing.

¹⁰³ Lg. 17.90 (1941), where Hitt. ḫullaš- ‘be defeated’ is nevertheless compared with Gk. ἁλίσκoμαɩ.

¹⁰⁴ RHA 4.274–5 (1938).

¹⁰⁵ Cf. Mélanges Grégoire 2.544–7 (1950).

¹⁰⁶ Pedersen, MS 50; Couvreur, Hett. Ḫ 113; Sturtevant, IHL §37d. Cf. Hitt. anda ḫark- = Akkad. kullu ‘enclose’.

¹⁰⁷ Cf. Couvreur, Hett. Ḫ 24, 111; Pedersen, Hitt. u. IE 126. Benjamin Schwartz, The root and its modification in PIE 22 fn. 23 (Baltimore, 1947), adduces Hitt. ^DUGḫarḫarān ‘a vessel for beverages’, originally ‘holder’, as an argument against the assumption that -k- is part of the Hittite root.

¹⁰⁸ Hethitisches 2.45 (Leipzig, 1922).

¹⁰⁹ Ibid. Cf. Sturtevant, Hittite glossary 21 (Lang. Monograph 9, 1931).

¹¹⁰ Cf. Sturtevant, HG² 5, 79 fn. 68.

¹¹¹ Friedrich, Staatsverträge des Ḫatti-Reiches in hethitischer Sprache 1.18, 20 (Leipzig, 1926).

¹¹² Helmuth Th. Bossert, Asia 58 (Istanbul, 1946); see also Sommer, Heth. u. Heth. 43.

¹¹³ Sturtevant, IHL §37f.

¹¹⁴ Friedrich, Staatsverträge des Ḫatti-Reiches in hethitischer Sprache 2.144 (Leipzig, 1930).

¹¹⁵ AOr. 17, 2.132–3 (1949).

¹¹⁶ André Vaillant, Revue des études slaves 26.141 (1950).

¹¹⁷ RHA 2.205 (1934).

¹¹⁸ It is not even quoted by Benveniste, Origines 1.162 (1935); Pedersen, MS 50 (1934), Hitt. u. IE 146 (1938); Sturtevant, IHL §56c. Couvreur, Hett. Ḫ 141–2 (1937), restates Evald Lidén's connection of Arm. harkanem with Arm. orot ‘thunder’ (Armenische Studien 88–90 [Göteborg, 1906]). from the PIE root *per- ‘strike’ (cf. Lith. Perkúnas ‘thunder god‘); but as Pedersen points out, A concise comparative Celtic grammar 387 n. (Göttingen, 1937), Arm. h- need not reflect PIE initial *p-.

¹¹⁹ Bergsland, RHA 4.275–6 (1938).

¹²⁰ Cf. Sturtevant, HG² 57.

¹²¹ Cf. Austin, Lg. 18.22–3 (1942); Sturtevant, IHL §22a; my survey, Mélanges Grégoire 2.539–40 (1950).

¹²² Lidén, op.cit. 87: ‘Die verschiedenen Gebrauchsweisen stimmen im Ganzen ... genau überein.‘

¹²³ Ibid. 52–4. Arm. u < *ō presumably reflects the characteristic lengthening in original root-nouns (PIE *xór-dh-).

¹²⁴ No better is the comparison with Skt. irasyáti ‘is angry’, still advocated by Julius Pokorny, Indogermanisches etymologisches Wörterbuch 337 (Bern, 1950).

¹²⁵ A similar interpretation of the Greek word was suggested by Louis Deroy at a meeting of the Cercle belge de linguistique (December 1950). It implies a different explanation of the assumed cognates; thus, Skt. ŕkṣaḥ ‘bear’ would be derived from the zero grade of PIE *xer-g- (IE -ḱ- on account of Av. arša-?) with the suffix -so-, probably of desiderative origin.

¹²⁶ Vladimir Georgiev, AOr. 17, 1.281 (1949), would admit a consonant shift parallel to that of Hieroglyphic Hittite (e.g. -ḫa: Lat. -que).

¹²⁷ In concluding this paper, I should like to express my thankfulness to Dr. Szemerényi for many stimulating hints, duly acknowledged in the footnotes above.