A question arises as to what word in the Pahl. trlr. translates ahva(° dao). It has been supposed that ¢@khshdkthd venders it (of course freely), But I greatly doubt that opinion. Tikhshakihd, like taband, may properly express Jré = fra (see below). Or it may, like tébdénéd so often elsewhere, have reference to zastdiS, and be either gloss, or emphatic iteration of the idea in dastih (sic). Is ahva(°do) translated at all? If not, we may have a most significant confirmation of Roth’s view. Ahvd (or ahvd) may not have been translated separately because its force was expressed in Sranémam. Other- wise where was the translator’s first person taken from? The trlr. may have seen ahvd - - frénemné = ‘we two are praying’; and freely rendered Srandmam = ‘I am praying’ or ‘worshipping’. Or was mé(é) a sufficient ex- pression of the first person to him, as it may be also to us?
Fri was no unknown root to the Pahl. trlr.; see fr(dyd, fr(d)yo, and Jr(ydi rendered by dést®, we therefore conclude, as it is only possible to conclude, that the trlr. had some such reading as frénemnd (K.*) in his
5 ee. a or eee SN ee 2 Po a. ee ° ee eee he ac 7 ay
MSS. before him. Pavan gimédnik is most valuable, hampérsén? intelligently free (possibly ferasdbyd was divided ferasd. + ---). The gloss ‘more than the A.’ was probably occasioned by the false Pie of frinemnd.
The other glosses are all intelligent. Ner. slightly modifies frandmam by prabravimi. He refers azydo to az = ‘to drive’, as describing a cow of three years, and fit for use. Vind, like bard, gives emphasis to the sepa- ration of the wicked.
The Persian text reads fravdémam, but this is merely accidental.
6. I am very far indeed from vehemently repudiating the Pahl. trans- lation for vafas, vishipishnd. (It may be mentioned in passing that one writer is a little too hasty in casting a slur upon this rendering. He thinks that the Pahl. trlr. read vashdpishno, the corrupt Pazand form, and, as he supposes vashiipishné to resemble vafis, he would have us indulge in merri- ment at the expense of the trlr. for copying the form of the word; but it is very doubtful indeed whether the trir. was aware of any such lettering as vashtipishnd. If he transliterated at all, he may well have understood the syllable 27, which is of prime importance to the word.) ‘Ruin’, as the object under discussion in the entire chapter, is by far the better meaning for vafas, not so however in Y. 48, 9 where it is also possible, but not so probable. If I were to follow the example of recent writers in other cases, I should suggest two totally different words for the two places. It is well possible that the Pahl. trlr. is right in this place, and wrong in the other. Opinions drifted from ne document to another, and translations were some- times mechanically repeated.
I therefore compare the Vedic vdpus in the sense of ‘inspired truth’, root vd, vi, u (2) = ‘to weave’; cp. ufydn?; but I add an emphatic alternative’ in the sense of the Pahl. ca not discussing an etymology, except.to recall the fact that vap = ‘to cast’ might well convey the idea of ‘ overthrow’ or ‘scattering’ in an evil sense, the meaning ‘sow’ being secondary.
Vydnayd seems awkward; its meaning is however well indicated by the Pahl. followed by Roth and others. I should explain the etymology as vi + ad + nayd from nt = ‘to lead’; cp. for form, and for meaning, vinayd = ‘dividing’. The word is an instr. = ‘with discrimination’.
The Pahl trlr. does not render vyd@nayd in Y. 44, 7 as amat vabdanéd = ,wenn er tut (!). The exclamation point is not mine. There is no such mond as vabdénéd in the place; but one is not surprised at such mistakes, as the Pahlavi alphabet requires close study. Asiatic scholars should be on their guard against Haug’s rendering for aévd - ahti-vistd (2), wnam ~ vitam- possidens.
He renders ratu§ dominus here, and ratio in v. 2. The word is con-
crete in both cases. Ahi as a nom. mase, is quite important; cp. also daé(é)td as possibly in the same form. I am not aware of any exact Indian analogy.
The Pahl. trlr. rendered what now stands as @ in eévaochad(t), as if it were separated and more extended, by pavan piimman.
As he treats @.¢. d (?), @.¢. dnd, and @.e. donhd in the same way, it is only reasonable to suppose that he had donhd, or some other word meaning ‘mouth’, in every case before him, or else that he so corrected those meaningless (?) fragments. See just above for a justification of vishtipishno as a rendering for vafds.
The question of the woes of the kine is never lost sight of. I think that ‘possess a Lord’ in view of Y. 29, 8 is rather better than ‘consider a lord (as an authority)’, although the latter is more idiomatic as Pahlavi.
Ner. perhaps meant ‘the clearing up of the ruin’. He curiously renders ahtikih vindishno by svdmibhrishtasya (sic). It is hard to say whether he meant ‘the fallen from the lord’, seeing vindsishnd () for vindishno, or nistd (sic) for vsté. Or could he have had in his mind’s eye bhrijj (!) in the sense of ‘prepare’, ‘provide’? As the Pahl. is intelligently free as to the application of tatashé and thwd, so the nirmitavan of Ner. seems to show that he did not fail to recognize the perfect
7. The ‘mathra of fatness’ (hardly ‘of invocation’) was the holy reve- lation, which (indirectly indicated in the vohd vastryad Y. 29, 1) protected and prospered husbandry, supplied material for the offering, food for the cattle, and for the consumers (cp. Y. 28, 10 barethya - + + sravdo). Hvé urushaé(éjibyd = hvo(a)rushae(éibyd, a reading long since discovered by Darmesteter from the Pahl. trlr, is attributed by one writer to another who received it at third hand.
Eed, or (é €. d. vd) (sic) are parts of the same shattered word, or words, which we have in Y. 32, 16, and in Y. 47, 2, but in a still more corrupt condition; and they are invariably restored by the Pahl. trir. to a consistent form; that is, they are translated pimman,
One writer is actually not aware that an attempt is here made by the Pahl. trlr. to render the text; and in fact the mechanical order of the words is inverted. Some, following authority, read éedvd = avd = dvitm. This is a brilliant suggestion, but hardly tenable. Haug read éedvd as ydvd (?) = tempore (?).
The case is simply this. Ahura declares that he appoints Zarathustra (thwa) as the needed Ratu, who was to declare the provision for salvation, the voht vdstryd, the barethyd vaintyd sravdo Y. 28, 10, the ereshvd ukhdha Y. 28, 6, the mathré (rdddo) of Y, 28, 7, the mdthrem deditoi§ Y. 29,70
the sdsndo (gishatd) for which hudemém -- vakh(e)dhrahyd was needed, Y. 29, 8, etc. In harmony with this we have in Y. 31, 3, (hizvd) thwahyad donhd = pavan pimman followed by all (inadvertently by some). In Y. 32, 16 we have Zed né@ restored to its form donhd by the pémman of the Pahl. trlr.; so again in Y. 47,2, we have de@ nit: Hizvd ukhdhdis vanhéus zed nit (donhd; but see below) mananhéd, till finally (Y. 29,7 may have been composed later than other sections) we may have the seemingly senseless ée@ vd recon- structed once more, while, to show the exceeding great probability that the trlr. emends correctly, see the remains of the old error (?) still adhering to donhdé fully written in Y. 28, 11 mainye@us hachéd thwd éedonhé (sic) fol- lowing fré ma sishd -- vaochanhe(é). In Y.29, 6 we have the same conjecture apparently erroneously made which had been correctly made elsewhere several times; or else the trlr. reproduces, as often, an ancient Zend text differing from that which stands copied beside his own. As the trlr. had so often hit upon the truth, I think he correctly reconstructs éed@ vd (?) into donhd@, or some lost word meaning ‘mouth’ here.
But this last case leads me to inquire whether he may not, as to some of the cases, put us on the track of a new stem. H(e)d, occurring four or five times, might possibly be a corrupt expression for éndé here, = instr. of &e)na = dnd = ‘mouth’. The letter f elsewhere, more than once, repre-
sents nasalized long @ (a). Compare 3Ee REO” fue = ameshé spenté for ameshéng spenténg = ameshdn(s) spentdan(s); f was sometimes written, or miswritten, for the two letters }¢ blended into one by ancient custom
(cp. ¢ for im). * Surely this lies closer than the explanation of ¢¢ = yy = ¥-
If this conjecture is correct, (it is of course given with reserve) v@ would be exceedingly well in place as meaning ‘indeed’ here (cp. Sansk. v@). We should then regard the de)d. nd (Y. 82,16 and Y. 47, 2) as equalling énd n@, and not needing full reconstruction, while a(e)donhd would, on the contrary, show the later creeping-in of the form énd, and its blending with the text. At all events every fact shows that the Pahl. trlr. indicates the reconstruction of some word meaning ‘mouth’ in each of these cases where donhad does not clearly stand; and all commentators suggest emendations. The Pahl. read as an ordinary text would be rendered: ‘He bestows bounty on the worker’; but val before kardar may represent the verbal prefix ‘the worker to’, ‘the completing artificer’. Maretaé(é)ibyd may have been ren- dered with an eye to an etymology, ‘mar’ being seen as = smar, and the
allusion to ‘reciting priests’ is not inapt.
ae
The glosses are harmless but useless. Notice wkdgayatt = vakhshinéd ; go also elsewhere forms of vakhsh are rendered by those of kde. Spiegel boldly renders Ner. 4: ‘Ormazd vermehr?. I confess I am at a loss how to render either kag or kag as ‘increase’; but I am nevertheless strongly inclined to think that Prof. v. Spiegel is correct here. An author should be rendered according to his context, and the meanings ‘make appear’ and ‘produce’ come together in the word paiddyish = ‘production’. Ner. follows the Gathd more closely than he does the Pahl., although he is influenced by both.
8. That this strophe, or verse, is in immediate answer to the question in v, 7 needs not to be stated. Ahura igs the speaker in verse 6, with which this v. 8 has the closest relationship. This requires Mazddo as the proper reading in 6, and likewise makes it desirable to take that gen. in the sense of the dat.; (see the dat. in the sense of the gen. in Y. 80, 1, and elsewhere in the Avesta, as well as in the Brahmanas); or else, vice versa, it shows that Ashdichd stands in this sense of the gen. That Mazda, occurring at every point as the name of the Deity, should here be used as an ordinary acc. pl. = cognitiones, Spriiche (Haug), or an instr., and indeed in the sense of ‘from memory’, I do not regard as probable. (The rendering ‘from memory’ had long circulated.) The composer’s wording is obscure enough, but it is not probable that he should positively have intended to mislead his hearers in the Gathas. In the later Avesta we may have a play upon words. If the instr. is read at all, the only sense resulting could be ‘with the help of me, Mazda’, or ‘in accompaniment with me’ (?), and, ‘as my agent’, (‘I myself, as the source of revelation, really declaring the doc- trines’), but the name of God would not naturally stand in the instrumental.
As to the names Zarathustra Spitama, the last was a family name; the etymology is obscure. I should be inclined to refer it to the root of spayathrahya; (spt = ¢vi). The form seems a superlative, the meaning the ‘most august’. Zarathustra has only its last member clear, wstra = ‘a camel’; agpa = ‘horse’, and wstra, were common component parts of names. Zarath might mean ‘glowing’ and refer to a light colour;’ ‘yellow-camels’ might be the sense, as Frashaostra meant ‘fleet-camels’(?). The form might be the weak participial stem of 2ar = hri (cp. haritdcva), zarath = zarat, th = ¢ (irregularly), An erroneous opinion was, that Zarathustra = zar = eairt = ‘golden’ and tusht (?) = Tistrya = ‘the star Tishtar’, and so ‘golden- star’. Then Haug’s views were eccentric, zar = jar ‘sing. and ustra = uttara = weiter, and so ‘excellent-singer’. His later view is hardly better,
jarat = ‘old’, and ustra = ‘superior’. A man’s name should not be ex-—
plained from his distinctions, or his history, in cases where it bears a strong —