Roth has brilliantly solved the difficulty of hanar(e) by taking the traditional (sic) sense of the word as it occurs in Y. 47, 5. He renders c however (last part) wdhrend Vieh und Mann ihm nichts eu Leide thun. Pastus virahyd (?) adrujanté, as gen. absolutes. Hanar(e) may, I think, contain the same elements as sine. Haug rendered quae cogitatio (Gesinnung) «+ - (et quae cogitatio ejus sit) qui non vitam ullum (ullo modo) (?) destrust (7) ate Spiegel (as usual) better, but taking hanare, like Haug (of course inde- pendently), as from han = san = ein wenig (freely). The abl. term. of the a declension when followed by chd is written Gad(f)ché with much regularity; but I think that later cases (as that here where °aad(t)chd disturbs the metre) are imitations from occurrences where the metre required dd(t)cha to be spoken as dad(t)chd; see remarks on p. 407. The Pahl. pavan vin- dsishné may well be a simple enlargement by anticipation of the rendering padafras for maé(é)nis, or, again, it may be the original translation, pddafrds being alternative. Hanar(e), correctly rendered in Y. 47, 5, seems here. curiously mistaken, as I hold, for a form of han = san, in its sense of ‘gain’, and it was so mistaken, as I think, owing to the following odd in vinasti, which was first explained to us as a form of vid by the Pahl. trlr.
here and elsewhere, an important grammatical datwm. The word usually transliterated nivekinédd may be properly mivédinédd rendering nivaé(é)- dhayé(e)mt, notice the New Persian naverd (sic). The misapprehension of hanare threw the renderings into disorder, yet see the correctness in lines a and c. The question arises, ‘is hanare translated at all’? I regard at as far from certain that pavanich mozd is not quasi gloss emphasizing nivékinédo (sic). We must seriously consider whether dard = ‘away’, ‘off’ in the gloss is not the echo of a lost rendering for hanare, so that in the inaccurate gloss we have a trace of the correct rendering of hanare, as in Y. 47, 5. Else where did bard shtkényén possibly come from? Even the limited idea of ‘announcing’ without the added inherent idea of ‘offered acquisition’ does not contain any hint of ‘freeing’, ‘separating’, ‘letting go’. Bard shtkényén contains additional translation as well as worthless gloss. (Of course we must understand in speaking of ‘traces of lost translation, or text’ that the Pahl. tris. have no certain connection at all with the Zend texts now written on the same paper with them, we might even say that it was ex- tremely improbable that the real trlr. had before him exactly the Zend texts which now appear transcribed in our MSS. beside his text. The copyists did nothing but copy; and the arrangers and compilers (editors) handed down Pahl. tris. from predecessors who saw MSS. hundreds of years older than those now extant. They worked these ancient opinions over doubtless, thus doing much mischief, and destroying much real tradition; see similar remarks elsewhere). As to Ner.’s Jamcha, see Spiegel’s Ner., also West’s Mainyo-i-khard II, 176 laviché = pédra.
16. A root d@ =*‘to know’ is not needed for huddnus, dé = aha, suffices for ‘the well disposing’, ‘the wise’. To Haug we owe the valuable rendering of yd-s(k)yaothanaschd as a compositum in the masc.; cp. yd-varend Y. 12, 7. This masc. may seem very easy and obvious at present, but that is the case with most discoveries. The Pahl. trlr., as too often in similar cases, when explained by the mischievous gloss, seems to doubt the aug- ment in asperezatd, committing the unpardonable offence of rendering it as an a priv. It is however always probable, or possible, that his text may be older than the gloss; and if that be the case, the text may have been intended to be read @kishidér = ‘emphatically a striver’, or ‘a striver against’; then the gloss only would be erroneous, and this supposition is imperatively necessary. If, however, the actual translator here really fails to see the augment, which is rarely expressed in Zend texts, he is yet, as ever, our first lexicographer as to the root meaning; he saw sperez in the sense of sprih.
17.. Haug’s velaman-faciat (celet) for aipidé-bdvayat (sic) is again rash
Bi
18. Dushitd, loc. sg. of °t to a *dubkshita, Otherwise, with Roth, cp. duriti, following the Pahl. through Ner. The Pahl. trir. renders sdzdiim _as if from sd@(g@) = ‘to sharpen’. The MSS. are in confusion as to the reading in a; J.3 sadbhyo, J.* bhavadbhyo, (so I think J.4 means, aS a correction of sadbhyo in J. which was its original; see Y. 34, 7.) Ner., as should be well marked, renders alternatively, by vighdtaya = ‘smite’, ‘kill’, and, in the gloss, by drachayata (so Haug) with the Pahl. The reading drayet (J.3) might however point to dr as a verb of motion in the sense of arpdyet = ‘send out’, ‘let fly (a blow)’. I note this, as Haug’s text of Ner. is so excessively erroneous that there is danger in trusting it; see his own apology for it. J.4 has here what seems a corrected dsdrayet, or dchar°.
19. Ahébis, or ahtimbis. Ahitmbis, ahtim.bi§ stand in the MSS. here (K> has aha.bis at Y. 44, 16). Ahdm was probably a separated ahu, provided by the early scholiast with the case sign of the acc. Cp. the case |
signs elsewhere before the suffix in yavaé(é)tdité(2), avaé(é)tds (°tdts). The Pahl. trlrs., who have given us such rich emendations as well as decisive render- ings, might be trusted here, as they were by Westergaard. But ahdmbi§ of the MSS. must be considered. As = ‘healer of the world’ it sounds rather advanced for the document, although thoroughly possible, Justi and Spiegel having long since pointed out d:§ = ‘healing’ in vispddcs, hubis, eredhw(v)dbis (names of plants). Ahdmbis is grammatically possible in some
places, but it is not probable in Y. 44, 16 chithrad méi dam ahibis ratim
chizhdi, where ahibisem would be required, unless the word were applied to Ahura, but ‘healer of world’ would apply rather to the prophet; also its position between daém and ratim does not favour a voc. AhimbiS = *asu- bhiksh = ‘seeking after life’ (for the people) might be considered. But the word as = ahibis means for ‘the people’ (so, long since privately circulated), or ‘the worlds’. As to the pl. apparently for dual., it may be said that no dual was in view. Ahvdéo = ‘of two lives’, but ‘the worlds’ might include Heaven, Earth and Hell, or the hamistakdn (?) might constitute a 3"4 state; see also ahvdhi Y. 51, 9. As to instr. for dat., cp. dregvddebis Y. 29, 2, Y. 48,11. There is of course a syntactical difficulty in holding erezhukhdhdi to be governed, as would seem so natural, by g@stéd (which governs the dat. of the person and the accus. of the thing); we should not expect to find the nom., which is quite natural before erezhukhdhdi, emerging far off in khshayamané, with erezhukhdhdi referring to the same person before it. I had therefore severed erezhukhdhdi from gistd@, and regarded it as ex- pressing the object of the energetic khshayamand hizvé.vasd, ‘the person’ here, but preferring ‘the thing’ in §.B.E. XXXI. But one does not know what to expect in the Gathds. Perhaps the meaning is: ‘Let him listen to the truthful speaker, who conceived of Asha, etc. - - - powerful over words, and free of speech’ - - - (2). Of course this is greatly to be preferred, if it is possible, and it may not be impossible, Haug makes the vidvdo the hearer (?), Spiegel (far better) takes the middle in the sense of the passive: ‘let the wise A. (but voc. for nom. (?)) be heard- - - who rules with truly spoken words, who has power in his tongue’, explaining; ‘he needs only to
‘speak, so it happens’. Haug: Vachanhim khshayamané hizvd.vasd =
‘dessen Zunge frei sich bewegt’, Roth gives an admirable suggestion that gastéd is understood; ‘(héret) auf seine Versicherung. Der Worte ist er miachtig, beherrscht seine Zunge’. De Harlez: ‘qui parle en maitre’, Jack- son; ‘who will have power at will over the words of his tongue—so that they will be verified - - -’ (see Spiegel). I prefer to regard the words as affording the reason for the admonition; the yé mantd ashem is declared to
have absolute command of speech, to give the infallible and inspired direc-
tions which are to be heard. Hrezhukhdhdi (S. B. E. XXXI, p. 51, infin. (?)) is concrete; cp. the eresvachdo contrasted with mithahvachdo v.12, But who is the vidvdo - - - khshayamandé hizvd.vas6? Almost everything points to ‘the prophet’ of the ‘holy Faith’, as opposed to the dregvant of v. 18, save only v. 7. Mazda also needs to be accounted for, and cannot well be termed an instr. (It is however possibly in place as a prayerful exclamation not affecting the syntax). Yas td mantd - -- Ashem (v. 7) almost forces us to refer the same words here to the same Divine Being whose activity had already been expressed by them in an earlier verse of the same hymn. Able writers, who have bewailed the obscurity of the Gathds, regard it as the critical procedure to seize every expression that bears any resemblance to language applying to the human subject, and to force all grammatical analogies, and parallel passages, which refer the words to the Deity, with one sweep out of the way. I myself also certainly hold that an alternative translation should be rendered in all cases where the words fit a human subject, and I also hold strongly that the text should be slightly changed where indicated, and alternative renderings afforded in emended texts; but the question is as to our jirst translation. I hold that the frequent inten- tional obscurity of the Gathic composer, or composers, should be critically held in view, and oblique address in the 3*4 person should be recognized as possible when made to Ahura, or a description in the 3'4 person made, with an address in the voc. in the midst of it. Can then yé mantd Ashem be referred to Mazda with a voc. Mazdé in c? See him referred to in the 34 person in two lines of v. 7, and in the voc. (224 person) in the 3*4 line. Then in Y. 32, 16, according to a disinterested criticism, yé(e)hyd md aithischid(t) dvaé(é)thd may well refer to some superhuman power; yet we have in this same Y. 32, 16 vocatives, Maedd Ahurd, thrown in. See also the curious structure in Y.44,1, and the voc. Mazddé in Y. 44, 2; see especially Y. 45, 4. I therefore conclude that the Deity was addressed in the Gathas sometimes with affected and intentional obscurity of diction, and that in view of the strong expression yé@ mantd- - ahtibis vidvdo, Ahura Mazda may be here indicated (possibly as representing his prophet). But I would add a most emphatic alternative in the other sense. Véddtd might well be loc. of a vidditi, but it seems needed as in agreement with dthrd. At first sight vanhdu seems loc. with viddta, but see the parallel Y. 47, 6 dthrd vanhdu vidditim (or °té(é)). Vanhdu is, like usta, a loc. used adverbially. The form — of gistd perplexed the Pahl. trlr., as it has others. But I rendered him | formerly far too much in the sense of ordinary Pahlavi. Nydkhshishnih : (practically = nydkhshishnd) is one of the frequent infinitives for imper. — (see Ner.’s grotavyam), indicating ‘propriety’, if not ‘obligation’, As ordinary ,
Pahl. I would render as formerly: ‘He whose listening is the hearing which is discretion and righteousness - - -. (6) The authority of speech rightly spoken, etc’. Vanhdu in c seems not rendered. Atrid was formerly read and rendered ‘pure’ by me from excess of carefulness; better diréd (?) = ‘the thoroughly impure’; cp. Pers. ridan, still better dérikht (so DJ. in Y. 51, 9, and probably also here) which might be intended to express @ ++ rikhta (irich = rich); or again we have had West’s agirdid = ‘undecided’; see Essays, p. 353. Ner. and the Parsi-pers. afford the proper meaning. Bikht (= ‘certain’ in the sense of ‘holy’, a frequent association of ideas) being in antithesis, our word, whatever it may exactly be, means ‘unstable’, or ‘unholy’. I do not follow Haug’s dhuvandvayor (sic); see recurring forms of dvaya.
20. As to the reading dyad(t), I agree entirely with Roth that déydd(t) is the allein brauchbar. The rendering ‘who comes over to the righteous’ I regard as very improbable.