ing was evidently in the first place 'cram, press together', then 'pack up, load, carry', and finally 'have'. If the loss of the final k before consonantal endings was due to a regular phonetic change, we can explain its retention in harkzi 'he is destroyed', etc., on the assumption that an original g (Lat. frango, frēgī) was retained in this position when original k was lost.

It seems probable that harandan (not harankal), which is explained in a glossary by Assyrian BIRTU and HALSU 'stronghold' is a Hittite word; and if so it is accusative from a nominative *haranza. This form would seem to be a participial stem from har (k)-, so that the original meaning was 'held, possessed'. I am therefore inclined to think that the distribution of stem-forms in the verb-har- before consonants and hark- before vowels-is not original, but that there has been a contamination of two different stems, one with and one without a final k. Possibly the stem har- is to be traced to IE *bher-- 'carry' (❤épw, fero, etc.) Probably another derivative of the root *bher is to be recognized in harnaus 'birth-stool',46 which, in that case, contains the causative suffix nu, and properly means 'causing to bring forth'. WaldePokorny, VWIS 2. 153 (following Persson), show that the root meant 'bring forth' in Indo-European as well as in Germanic. Sommer and Ehelolf think that the many variant forms of harnaus indicate a loanword. If they are right the source is probably Luwian, and that would not interfere with our etymology.

An important group of Hittite words consists of handa- 'fix, determine, construct', handaš 'true', para handalar 'the power by which gods determine the fate of men, numen', para handanda- 'rule by divine power', and para handandatar = para handatar.47 With the stem hand- we may compare IE *bhendh 'bind', Sanskrit bandhati, Avestan bandaiti, Gothic and Anglo-Saxon bindan, Greek Tevde pós 'father-in-law', Latin offendimentum 'band by which the apex was held in place'. As to the final consonant of the root, we should expect IE dh to appear in Hittite as the sound which is written indifferently as d or t. For that is the correspondence already known in dai-, tiya-, te- 'place', which is certainly cognate with IE *dhe- 'place'.48 See below p. 121.

Another word with initial h corresponding to IE bh is hanna- 'litigate, decide a law-suit', whose derivative hanneššar means 'justice, court of

"See Sayce, ZA. 4. 386 (1889).

46 For the meaning and forms, see Sommer and Ehelolf, BoSt. 10. 3f. 47 See Sommer and Ehelolf, BoSt. 10. 30f.; Götze, Hatt. 53f., 60, 91ff. 48 So Friedrich, ZDMG, NF. 1. 159 (1922).

justice'.49 This is Greek paivw 'show', Albanian be„j, beń (<*banyō) 'make'. This root is an extension of *bha-, the base of Sanskrit bhāti 'shines', so that hanna- is ultimately akin to *haššuš 'sun, king'. For the present we must leave undecided whether the nn of hanna- resulted from the ny which is evidenced by the Greek present stem, or whether it is an inexact writing for the single n of ἐφάνην, φανερός, etc.

Friedrich, ZA NF. 1. 189, has clearly shown that hurtaiš is equivalent to Assyrian IRRITU 'curse'. It is quite possible, however, that the word sometimes had a more general sense, such as 'harm, destruction'. K Bo. 2. 3. 2. 1f-50 nu ku-u-uš ENMEŠ ZÚR i-da-a-la-u-e-eš hur-da-a-e-eš QA.TAM.MA li-e ú-wa-an-zi may very well be translated 'just so may not evil destruction come to these sacrificers'. The verb hurt- in the military oath has as its subject the gods of the oaths, li-en-ki-áš DINGIRMES, whose function is to destroy, not to curse, the man who violates his oaths.

If the primary meaning of hurt- is 'injure, destroy', we may connect it with Greek Téρłw, Tорłew 'destroy', Skt. bhartsati 'threatens', from IE *bherdh-.52

The verb halza- 'call, recite', whose forms were discussed above (fn. 31), is clearly IE *bhels-, from which come Skt. bháşati 'barks' (earlier *bhṛṣati), bhāṣate 'speaks' (earlier *bharşate), Lith. balsas 'voice', bilstu 'I begin to speak', OHG, AS bellan 'bark', etc.53 I cannot cite another instance of the change of ls to lts; in fact I do not know of another Hittite word which contains the sound group lts (l-z). But the change is so closely parallel to that of ns to nts (e.g., in the nominative singular of the nt-stems), that no one will be inclined to doubt it. Some difficulty is presented by alšanza 'captive?,54 but we can safely leave the explanation until we discover the etymology of the latter word.

The adjective harkiš 'white' is cognate with Skt. bhrājás 'gleaming', Alb. bare 'white', Goth. baírhts 'bright', and Lith bérszt 'grows white', whose IE base is *bherēĝ-.56

Sommer's discussion of the verb hatk- (BoSt. 7. 1-6) clearly established

49 See Sommer, ZA 33. 931 (1921); Friedrich, OLZ 26. 45f. (1923).

50 Transliterated and incorrectly translated by Hrozný, BoSt. 3. 72f.

51 KBo. 6. 34. 4. 12; transliterated and translated by Friedrich, ZA, NF. 1, 168. 52 See Walde-Pokorny, VWIS 2. 174.

53 See Brugmann, Grundriss 12 430, 459, 778.

"See Friedrich, ZA, NF. 2. 274.

"See Friedrich, ZA, NF. 3. 185.

"See Walde-Pokorny, VWIS 2. 170.

the meaning 'shut' (a door, a temple, etc.), and showed that another meaning such as 'drücken, bedrängen' must be assumed. Sommer connected the two meanings by citing an Accadian parallel, and suggesting that the latter language influenced the Hittite verb. To me such foreign influence upon the meaning of a native word seems rather improbable, and besides there is reason for making the second meaning of hatk-somewhat more drastic. The words hatugais 'frightful', hatugatar 'terror', and hadukišzi ‘atrocities are committed' (in a hostile country)57 can scarcely be separated from our verb; and accordingly I would translate anda hatkešnumi 'I cause atrocities to be inflicted in (a hostile land)'. The primary meaning was probably 'strike, smite'. For the development of the meaning 'close' (a door), compare German zuschlagen. Without the suffix (u)k we have the iterative-intensive hazzik- (i.e. hat-sk-) 'strike, play' (a musical instrument).58 If Sommer (BoSt. 7. 57) is right in interpreting hazziya- as 'strike, engrave', that verb also belongs here. The Hittite root hat-(uk)- is probably to be connected with the IE root *bhat-, which is assumed by Walde-Pokorny, VWIS. 2. 125 f. for Lat. fatuus (*mit Dummheit geschlagen') 'foolish', Gallic-Latin battuo 'strike', Russian batu ‘oaken stick', etc. Note the extension in u in Latin and Gallic as well as in Hittite. It is possible instead to think of a connection with Skt. bādh- (from *bhādh-) ‘harass, distress', which in that case must have undergone a development parallel with that of the Hittite verb.

Hittite haršan 'head', whence haršanallis 'wreath',59 may be connected with Skt. bhṛṣṭiş 'point', Lat. fastigium 'peak, summit', from IE *bhares- 'point, tip'.60

61

With some hesitation I suggest a connection between has 'soap', whose original meaning was probably 'ashes',"1 and Skt. bhásman ‘ashes'. The latter word is connected with bhas- 'crush, chew', etc., and so the suggested etymology carries with it the assumption that Hittite haš originally meant 'that which has been reduced to dust'. There can be no doubt that haššaš 'hearth'62 is connected in some way with has; did it originally mean simply 'ashes'?

Since IE bh appears so frequently in Hittite as h, one might perhaps

57 See Friedrich. ZA, NF. 3. 189 f.

58 See Götze, Hatt. 101 and fn. 1.

59 See Götze, Hatt. 953; Friedrich, ZA, NF. 2. 275.

60 See Walde-Pokorny, VWIS 2. 131.

1 See Friedrich, ZA., NF. 3. 191 and fn. 4.

62 See Sommer and Ehelolf, BoSt. 10. 243.

=

=

expect to find this the regular development of all the IE voiced aspirates; but this proves not to be the case. I have already mentioned two words in which IE dh appears in Hittite as the sound which is written either d or t; namely, dai-, tiya-, te- 'place' IE *dhe, and handa- 'fix, determine' IE *bhendh-. A similar development has to be assumed for the labio-velar voiced aspirate on account of Hittite kuen-, kun- ‘strike, kill' beside Gk. Oeivw, Skt. han-, etc.63 The pure velar aspirate appears as g in dalugasti 'length', which is the same word as Church Slavonic dlugosti, while its adjectival base is equivalent to Gk. doλixós, Skt. dirghas, etc. I would connect kesr 'hand' (ki-eš-šar, dat.-loc. ki-išri, ki-iš-šá-ri, etc.)65 with Gk. xeip, and thus explain the troublesome forms with xeup-(xnp-,xepp-) on the basis of earlier xeap-. The stemform yep- may have arisen in primitive Greek before consonants (xépvių) and by dissimilation before suffixal σ (-xepns from -xeo pms). If we thus separate xeíp from IE *ĝher-, it may contain either velar or palatal.

Apparently IE bh also appears in Hittite as p in the interior of a word; for it is impossible to separate nepis 'sky' from IE *nebhos 'cloud'. It follows that in several words discussed above the medial h was due to the analogy of an initial h in related forms. Thus miyahuwanza 'growing old' was influenced by the verb huwa- 'grow', and the reduplicated huihuiššuwališ got its medial h from the verb huiš-.

There are, however, many words with medial h that cannot be explained by such an analogy. There must be some other source or sources of the sound. Without attempting a complete investigation of the various possibilities that will occur to anyone, I want to suggest that in several cases Hittite medial h represents an original sound which has been lost in Indo-European. The Hittite word for 'fire' is pahhur, which appears also in a more primitive form pahhuwar, genitive pahhuenas.67 It is therefore an r/n-stem, and it must be connected with the IE r/n-stem which is seen in Gk. Tuρ, Umbrian pir, Anglo-Saxon fýr, beside Goth. fōn, Old Norse funi, Old Prussian panno, etc. Here as elsewhere I am not yet certain about vocalism; but if we may tentatively use the vowel e (somewhat in the manner of the Egyptologists) the Proto-Indo-European word may be given as *pehewer/n.

"See Friedrich, ZDMG NF. 1. 159 (1922).

64 See Hrozný, SH 23.

"See Sommer and Ehelolf, BoSt. 10. 34 and references; Ehelolf, Indogermanische Forschungen 43. 317.

66 See Hrozný, BoSt. 3. 725.

"See Friedrich, ZA, NF. 1. 188, ZDMG, NF. 1. 159.

68

A similar treatment is suggested for the stem seen in lahuwa- 'pour' and lahhus 'basin < into which water is poured in washing the hands>'." If we remove the medial h we have left the consonants of the IE base of Lat. lavo, Gk. Now, etc.

Medial h appears in several personal endings of the second or hiconjugation and of the medio-passive voice. If these endings were inherited, Indo-European must have lost the h here also; but a final decision seems impossible until we know more about the Hittite vowel system.

UR

Postscript: With huwa-, hui-, huis- 'live, grow, beget' we must certainly connect huwaši 'statue,' whose ideographic representation is ZAZI.KIN (Sommer and Ehelolf, BoSt. 10. 11). There has been some hesitation about the meaning of the word (Hrozný BoSt. 3.85; Friedrich AO 23, Heft 2, Nachträge 5*); but it seems to be fixed by a comparison of K Bo. 4.10.1.20-21 with Hatt. 4.72. The former passage (on which cf. Forrer, Forschungen 1. 7) runs: Kur-šá-wa-an-šá-áš-ma-kán (21) EGIR MUH UR.KU GÁL Zhu-u-wa-ši ZAG.áš, 'Next, however, back, above Kursawansas the big-dog-statue is the boundary.' There would be no difficulty in translating by 'relief' or the like here, although the Hittites are known to have set up huge statues of animals at various places; but when Hattusilis records the houses and estates which he has given to Ishtar and adds: hu-u-ma-an-ti-ya-mit EGIR-an ZAZI.KIN ti-it-ta-nu-uš-kán-zi, we must translate 'In all of them <men> will again set up the <cult> statue.' The word therefore means primarily 'representation of a living being,' a meaning which is equally appropriate for the ideogram.

68 See Weidner, Archiv für Keilschriftforschung 1.12 (1923); Sommer and Ehelolf, BoSt. 10. 74; Friedrich, ZA, NF. 1. 11.