16. Notice the difficulties as to caesura ind. Pronounce Huwégwa(?), or Huvdgrd (2), at all events with three syllables. With some wstd stéi is reduced to ‘of whom we two wish that it may go well’ (sté¢ infin.). Ustd is however the hosannah of the Gathds, and really an adverbial loc. as inter- jection, ‘salvation!’, also I think sometimes used as we might use ‘salvation to a people’; that is, ‘salvation-bringers’. I5ta, according to a very old suggestion long circulated, would be a loc. of #sti; but if khshathrd is read with K.5, the word is a pl. of the participle. As I have said in the other work, 7@st@ khshathra might cover a var(e)demam = °man, as a loc. in my former alternative sense of ‘choice abode’. Cp. also Y. 37, 3 tem ad(t) ahiirya nédment mazddvard spentétemda yazamaidé(é); cp. the name bijavara (the best of the grains), also the names Buddhivara. Bréhmanavara; cp. also the throng of words with the first member vara. So much for my former alternative; my preferred one is, ‘in fulness of power’, to vridh (vardh), loc. of var(e)dman (two syllables); others ‘in the glory’, also ‘in the citadel’. As altern. trl. with the reading khshathrem as nom. sg., render; ‘where the kingdom is in the possession (?7s¢@) of the good spirit’. Hastishn reproduces stéi as if an infin. The root idea only of usvah?, whose rare form does not seem recognized, is rendered by khérsandth in connection with a 224 s¢. Read Huogudjah in my former printing. Tam prasddam must be an accidental mistake. I should still prefer to correct to a nom. Ipsdno (so) renders khvdsts yekdviminéd as a pert.
17. Caesura at the 4%, yet notice the difficulty at a Read my former printing [secundum recte] comparatas - - redditam comparatae sunt. _ For afshmdné the sense of the Pahl. is nearly universally accepted; possibly ap is the root, and the idea of ‘elaboration’ is expressed. Metrical com- position was sacred to the Deity, notige the use of dpas as expressing the sacred functions of the altar. The word may be *apasman, loc. with weakened stem; anafshmam = °man is also loc., cp. Vedic occurrences of loc. in an. Sénhdni may be also loc.; an acc, pl. nt. is also not impossible, Y. 53, 6, or, lastly, the word may be a 1 sg. conj. The suggestion has been made that anafshmaim may be a gen. pl. (?). Some scholars might be disposed to regard vista (so) as a 274 pl. = *vitta (so): ‘Did you ever obtain’. One writer prefers reading véstd vé(s) + ta (cp. yergstd); ‘with this “Your” obedience the pious praises’. Vis = ‘to come’, ‘to be’ has been thought of, hada vista (s0) = ‘together being’. Vistd-vahmeng might well be a compos. ; but vistd (so) is not impossibly an instr. with Sraoshd ‘with accepted, i. e. re- cognized, obedience’. Our translations should be based upon ou judgment,
and not sacrificed to a wish to exhibit fresh but improbable possibilities. These should be. presented, if anywhere, in notes. One writer prefers mantt aS noun of the agent, ‘with his clever (dangra) ‘observer’, Asha’. I prefer Ashé as adverb. instr. abstract, although, as ever, with latent personality. Sraoshé has been regarded ag a 1st Sg. aor. conj. (srw) by some, ‘that I might hear’ (?). Can manta possibly be a verbal form in the active (Indian mid.)? Differences between Indian and Iranian equally great exist, and the connection temptingly calls for the verbal form. As to pavan aito-dahishnih formerly rendered literally with the sign. of warning, cp. p. 413, and read ‘in the continuing existence’, Padméni for marti is followed in my former alternative sapientia. Ner. at least testifies nega- tively against the acceptation atts as — ‘is’; sahaddtyd = aitd-dahishnth. Dakshini as a formation is possible, and may well have been meant. Read my former printing Huogudjah. :
18. Read my former text vichithem. Some writers, observing ydus the aorist of yw, have suggested; ‘who unites himself with me’ (?), Yao§, like the Indian yés, is indeclinable, we might take it as nom., or ace. Read as alternative: ‘Who is holy (?) toward me--., As chid is found in the Indian after verbs, ep. RV. 135, 9, it looks the less strange in aschtd(t), or can as = ‘verily’ (?) + chid(t), Strengthening? Asd@ (So) we Should naturally suppose to be a nom. of astar = ‘oppressor’; but it is not impossibly an acc. pl. Daidéta as gra Sg. mid. opt., we should naturally render ‘who may accept us’, that is, ‘treat us’ as an oppressor’. But in the Gathds, as well as in the Indian, we are often forced to accept sub- stantially an active sense for the middle of dé (= dd, or dhé); see on p. 510. And it is a question whether the purposely varied accus. pl. in astd, if the form be such (?), may not have literally as well as implicitly the sense: ‘I will send affiction upon him, who may send afflictions upon — us’. This Suggestion is however alternative. Reading @stdi, with other MSS., we have: ‘who may hold us for oppression’. Véchithem is better in view of Y. 30,2. Read my former printing . . [injuriam] - +; [id]. +». The Parsi-pers., reading jan, renders ‘life’ (not so in Y. 28,0 where yan = yanim). Ydn accords better with yaos. Pavan tani (tand) = ‘in reality’, — freely for aschid(t); or for some form of ast?. Chdshédo, so according to the gloss is difficult, Dropping the gloss, chdshidé (same signs, cp. chdshidar), — might equal ‘I taught’. Andstth expressed, or was associated with, oppres- sion, yet Gsté read as anst(i)h may have suggested the form chosen. Pahl.
» = h or a, or a; 4EN)ss = ansid or anastih (sic) (not andstih). The
gloss diné (so) mars e, Read my former printing ydn, sadai ’va.
19. Caesura at the 48; disregard all former divisions after the 6t syllable as wholly informal, and intended to mark for the non-specialist a stress of voice on the 7, Mané.vistéis = mané.vistdis must mean mente- comparatis, or something similar. A valuable suggestion, long privately circulated, was ‘with all that the heart can desire’. We might consider the reading vistdzs (so) to vis = ‘arrived in the mind’, or véstdi§ (so) might equal vi + stdis = ‘pervading (extended in) the mind’, to std, root declension with transition to a. I formerly thought that the change to ménd might be preferable. See the Pahl. Also md niévistdis (see the MS. K.4(?) mani) might be considered; ‘together with all things announced by me for the sake of (instr. in the sense of ‘concerning’) the mother kine’. One scholar would render gavd az? as duals, referring to the later use of the expression azt in the Vend. I think that the ah? gé was the same whose soul bewails in Y. 29, 1. Of course all other documents are inferior to the Gathas on this point. The Indian af? finds its explanation here; see also above, p. 421. One scholar would refer s#s. to sad = ‘execute’. I was formerly inclined to take sas as the nom. of the root, but it is better to take it with Spiegel, Justi, and Bartholomae (in A. I. Verbum) as the usual sg. with loss of S. ‘Thou hast declared to me, O M. as the most wise’; see above, p. 516. Read my former printing mente-comparatis - -, - - nuntiabas |- ties, -abis| - «. The Pahl. trlr. seems to have read ménd for manz. His ‘announcing’ must be understood in the sense ‘inviting by announcing’, and so ‘acquiring’, ‘gaining’; see Ner.’s grihituh (sic). The root vid I think is present, even in the form nivékéd which I consider a corruption. For sds he probably read vas. Sas is correctly rendered by gift in Y. 43, 11, whereas forms of vas are freely rendered by Combinations with khdirsandth, as the root of usvaht is correctly seen, and freely rendered in v. 16.
Gatha(a) Speitamainyiai(u). Yasna 47.
For introduction, summary, and further comment see 8. B. EK. XXXI, pp. 145—147.
1. Caesura after the 4, yet see c; stress on the 7% syllable. Ahmadi = ‘to this one’ refers to some subject in a lost verse, cp. the ¢dthréy(v)a of Y. 43,2; see ahmdi in v.3, ashaoné(é)(?) v.4, ashdunéé) (sic) v.5. Other- wise it equals ‘to us’; exegetical difference nil, ‘to a representative saint’ = ‘tous’. Ihad formerly preferred regarding da@(dan) as the pres. part. from the stem dé in mutilated Vedic form; that is, with no sign of the nom.,
and in agreement with Mazddo--- Ahurd, or ag 3ra pl.. of improper conj. aor., agreeing with the subjects involved in the instrs. + Ahuré. The verse is purposely and artificially crowded with the names of all the Ameshéspends (Vahista Mananhd being of course only a variant for Vohi M.), hence the possibly pl. form. Some hold to an acc, infin., leaving the strophe without a finite verb, Read Armaéti in my former free trl., and Sapiens as alternative in all occurrences of Magni-donator. The glosses. alone prevent a closeness in the translation. Many follow améné who do not adequately estimate the Pahl. elsewhere. As per contra, see ahmdi in. v. 3, v.6, and ashdwné(é) in v.5 which gathers up the sense of the previous verses. Mainyé was not mistaken for a voc. Read nikirido, khvéshan. Ner. renders ahmdi in the 1% pers. Mahdjnaninah does not accord with svdm?. Read °manyah. Mahattardt adrigyatvdt, poss. nigérend.
2. Hednd (2) = dnd na (2), or donha (?) = ‘with mouth’; see the Pahl.; with others, a corruption from anu which is referred to verezyad(t): see on Y. 29, 7. Oyd = aya (Spiegel), otherwise — avayd. Pahl. The gloss induces an awkward separation of d. Zak 2 Jarzdénak is corrected (as to form) by pavan dandkih which seems to have rendered an instr. mazda not taken as the proper name. (N. B. the Pahl. trir. was the first author of this suggestion which some follow in other places as a brilliant innovation ; see Ner.) Read va piimman with D. and also DJ.s va ptimman, 224 occur- rence. I did not read it formerly because DJ. seemed to have no va in the first occurrence. D. supplies aoa é, substantiating my conjecture, but it is a modern MS. Perhaps I have erred in following P.’s correction to mukhena- cha which I did in view of the 2n4 mukhenacha, yet see the awkward loc. Jthvdyam, which probably occasioned the more sensible instr. in the gloss. J.*, although here very carelessly written, affords the excellent emendation vaktum for kartum in 6 gloss. Read; ‘what it is fitting to declare with tongue and mouth, he declares as the better work, or ‘in a more beneficent manner’. I had corrected to mirvdne with Sp., as I had only his MSS. gene- rously loaned to me; read nirvénajndne (J), also dahan, pédaish, or paiddyish,
3. As improvements since my former printing, read mainyeus, skereitim. Read as alternative cle for ila, and ‘he’ for ‘she’ in my former printing ; (@) ‘Since he with good mind’s wisdom counsel taketh’. I would now prefer referring hem - frastd to the subject in ahmdi (unless rendered ‘to us *, ne which case the Cow (Kine, though pl., was used to avoid the poetic use of — ‘cow’) representing the tribes, would be the questioner, as in Y. 29), ‘To — us’ is difficult with the sg. frastd. Here we have perhaps the origin of the mythical identification of Aramaiti with the earth. Ahura appointed Piety
as the ultimate:cause of agricultural prosperity. But the sacred passage was afterwards rendered literally. Pahl. The sequence of words in @ proves accident, or freedom. The trlr. was not ignorant that ah? = hémanih; see Y. 32, 7; Y.34, 11; Y.51, 3; see Ner.’s asd. Ner. alsorenders ahmdi again in the 18* pers. Omit the hyphen after 6, which was a clumsy indication of connection. Samartham (so all) is a fair case to show that Ner.’s Sanskrit requires special treatment; here. ‘corresponding’, ‘good’, cannot. be the mean- ing. Nor is there any word in the Gathd which gives a clue. The word is here used in the sense of kold; yet some have supposed that Ner.’s text needs no translation. G'du with some as more orig., or gdv; ziydd.
4. Caesura at the 4; final «9, om. former misprinted‘?’. Regarding
the trl. above as secondary, read as preferred alternative. Ab hoc spiritu vulnerantur scelesti [-daemonibus servientes (dregvantd)|, O M. (leg. Mazda) benefico |spiritu], non sic sancti | fideles|. Kaxigqui|-ae| etiam |dignitatis| vir sancto ad gratificandum erat |sit vel erit|, possidens etiam ens magni |-nae dig- nitatis| malus scelesto-infideli, [ad serviendum est]. Free trl.: ‘From Mazda’s bounteous one, not thus the pure, Though feeble men alone would serve the righteous, yet for the wicked stands the great in power’. Rdres(k)yanti (so) must be taken as an intens. with subject named in a lost verse, or else for dregvatd we must read dregvantd with many MSS. ‘The wicked are