me with joy-inspiring help’. As an acc. pl. nt., cp. djushténi, djushti. For
‘come to my aid’, cp. us méi dr(e)shud, Y. 33,12, also the frequent Ved. ¢ gah. Vidé perhaps best = vindéd. The lengthened 7 compensates the loss of the nasal; as we have vinddti (a class) vindd may be 224 sg. imper., or vida may equal *xindd as 15* sg. conj.-imper. = °dni. In having formerly
printed da, I followed an opinion current at the time. It may be regarded © as a general expression for vidd = vindd as a poss. imper. = ‘obtain for .
me’; see the free ‘send to’; so, in view of gaéd?, and poss. imper. (d)rapd, better, ‘may I obtain’. Pahl. [Damdnd] = [yuge], and pafré, mistaken for a form of pd, caused disorder of syntax. Badtém may well hint aright.
6 is not exact, but recovers the general sense. Without gloss 7 shapir is
gen. = vanheus; ydmtininishnd infin. for imper. = gaidi; ardminidar, verb. noun for verb; rdminiddr yeheviindd (almost beyond doubt an exegetical gloss) = ‘it, the reading, may be rapa’ (so), khavitinishnd infin. for imper. Without gl. all but a is fairly close. Om. zit, c =: ‘Let there be
the giving of the good, a bringing of that which is a not-gladdener [it may be a gladdener (so the Pahl.)] (d) Let there be a knowing of the destruction’ etc. D. corroborates hamdi, aigham, ghal (a), dash-nikirdi b, eit i, aigh li (so M.1), (c)adds madam saryd after araminidér. Read dish-nikirét. Atra perhaps points to zak, otherwise antar with H. Correct my former misprint to
rakshém with the MSS. Haug was mistaken in censuring prdpnoti for arapd;
it renders ydmtiéninishnd = gaidt. Jidnatd refers adsh = aoshd to ushi (2) so = ‘enlightenment’ (?); see the Pers. Arapéd (= ardminidar, or rdm°) is replaced by Bahmanasya. Om. my former first*!. Read agh.
2. Read haesitaverim and Thkaéshé (for the antiquated D°) in my former printing. Rédreshd is hardly so probably a 274 g¢.; ‘may’st thou wound’; see the 34 sg. before, and the 34 sg. after. It is more probably a nom. sg. from the intens., ‘receiving much injury’; cp. Ved. rish; or =
‘wounding much’. Ménayé(e)it? is more naturally from man = ‘pause with
hesitating thought’. It is doubtful whether mamanyat (active), RV. X. 31,2 can be compared. Is the original idea of man ‘to stop’, hence ‘to think’? It is not likely that so abstract a conception as ‘thinking’ was original. Stoi seems dat. from std, cp. Y. 31, 8; with others infin. of ah = ‘to be’: ‘He has not maintained Piety to be (?) for as’. See Roth on Y. 31, s. 23, who prefers ‘for this land’. The Pahl. preserves only the root. Some might render: ‘The evil-minded spokesman of this connection makes me reproaches, the impostor, recreant from the Law. He tolerates no holy zeal amongst us, nor pious assemblies with the faithful’, Pahl. without gloss: ‘Thus me by his diseased impurity he causes to be retarded’. Vimdrth may shed important light on the nickname of the enemy. Méninéd could not natur-
ally = ‘cause to be anxious’ (man). Dérest was read from dares. Stéi may have been seen as dat. infin., freely = yekavimiinéd (see stéi = (h)astishnih in Y. 50, 6, etc.). D. has vdddndid hand ash 7 avi levind in a, afash in b gl., vddindnd in c, om. va before ldch, pavan fradranodth (no 1d) in d. Read digh, ldyik® (?); firtb(?), or fartb(?) might be read for fréb, as the usual Parsi is frév, I ventured on fréd as nearer Parsi. Ner. Drishtdntam points to maninéd as = ‘causes to think’. Nydyi is not amiss, with widened sense. Chhadmakah, cp. chhadman. J.4 has upari chhedah. Durdgachchhan occurs in Y. 30, 6 (there pl.) as = débdrédé what is the sense of it here?; see the Pahl.; Ner. referred dérest to dvdr (!), 6 may have stood in the Pahlavi character, = 0, or v with inherent a; hence the durdgachchhan, as di- badrastané goes back on dvar; Ner.’s mistake is most interesting. Read dgh.
3. As to caesura, see ahmdi and varendi divided by the formal pause. Read final eg5. D(t)kaé(é)shai is naturally ‘the evil lore’ (less probably ‘the
evil law-giver’) in antithesis with varendi. Saré = saréd, nt. acc. from sarah, ep. sarahsu, Yt. 10, 40; others take it as gen. with izyd. Antare - - mruyé(é) = °uwvé was first explained by the Pahl. trlr. Read ¢/ delectuc in my former printing. Pahl. D. has aétéindch zak 7 avi (a), ash 7 for aish (6), om. 7 after andarg (d). Read more lit.: ‘And thus, O Lord, for that religious desire he, or it, has been settled’, etc. Kdmak (+ am ait understood) renders zzyd well. The idiomatic ajitare mruyé(é) was first explained by the trlr. Read hamkhakih (so DJ., D. and M.1; two types formerly failed to act in - the press). Avabodham, too rigidly rendered ‘vigilance’, should equal ‘knowledge’, see varendi. I have corrected the jidyak of the MS. (with Sp.) after nydyt = dastdbar, see 7 dastdbar here corresponding to our -word. Sadai ’va is an error for hdmd?° falsely read in hamkhdkih, or possibly for a mistaken min hamishak (2); but see. the correct altern. trl. métratd° in . the gloss.(M.1 has pavan Adhar® zak in v. 2, d, D. dahishn dast6°(b), déstih L° here).
4, Rémem is better taken as personified with Aé(é)shemem, and we - cannot do better than follow the Pahl. as to meaning, and perhaps as to text likewise; reading remem, see Y. 48, 7, and Y. 29, 1. In Y. 48, 11 and Y. 53, 8 we have of course a totally diff. word = arméshtih + akdrih. Vas cannot be better explained than as Justi did so long ago as = vdichh (root noun), and following the Pahl. trlr. who recognized the word centuries before a Sanskrit word was known in Europe. Others seem to prefer the root van + s (?) in the sense ‘overcome’. Dan, to dhd perhaps better than to dd; ‘they establish’ the D.’. Some have suggested: ‘Whose evil deeds (duzhvarstd) one can never overbid with good deeds, these bring the Daé(é)vas their evil character’. Pahl. The gloss forces ‘upon whom’,
without it, ‘by whom’. Poss. vahrid(?) is better than vdrid = vareden. Aréskoch is better, elsewhere araské or °kun. The trlr. may have under- stood 6; ‘by their own tongues among the increasers of the decreasing one’. Read /shivintdané (D.). Dan is paraphrazed freely and intelligently. — D. and M.1, tye) ey for ty ey. Read khéshmo, or aéshmé. Ner.’s varshantah is figurative, but = vdréd; sammdrjanam hardly = aréskéch (?) but some mistaken form of @ priv. + résh, or poss. rich = ‘the not injurious’; sam- marjanam = virdyishnd = rdzaré, Y. 34, 12, = virdstand, 45, 3 gl.; sam- © marjayati, Y.29, 1 = sdzéd = sastd. No fshivinénd was (D.) supposed to equal vindganam etc. Let it be remembered that é) may = v, and vishdved = | vishoped, fshiiv and vishév might have the same characters. Ner. deciphered as a form of vishiiftand = ‘to destroy’, or pas, hénatan = ‘to scatter? might have been surmised in an ‘evil’ sense. Parsis of old had trouble with the shifting letters. Qubhoditd, as with hit-varsht, seems to show a reading hi-rost. Firib, or farib would be more Pers.; read ikhtiydr.
5. Read as altern. for huzeituS bonus civis, see above on Y. 43, 3, 4 and cum his omnibus, for the instr. as altern. As to line d, I now prefer my rendering of ¢di$ vtspadis in the sociative sense; see 8. B. E. XXXI, p. 164; ‘with all those in Thy kingdom, O A’. D. om. 7 after valman in a, ins. 7 after zak in b, read vddiindnd (so in c), has diff. sign for s» in kaddrchdi. Vddényén in the glosses may be imper.; see Ner. Glosses and Ner. aside, the trl. comes out nearer the Gatha. J.3 read kurute in } gloss, which seems to fit the text, but see the Pahl. and c. Kuru — vddinyén as imper. may have an alternative trl. in kurute (J.3, J 4, c) which is at least in the 374 pers. Satkdér°, generally renders forms of shndyinidand, and makes it pos- - sible, if not probable, that such a root was seen in what now stands as shindsakth. Shirtnt, so for shér°, .
6. As to caesura, see 6; read Sraés(k)yd. Read in me as altern. for mihi, and Vestri-devoti; in the free; That Insight of Your faithful one, O Lord. Can fraé(é)s(k)y4 = ‘now I declare forth’? The view that mraté(é) is an infin. had circulated privately for years; and also, so far as I know, © the objection that it stands at the end of the line was first suggested by a private remark of my own to a friend in 1882. Before that I had never heard, or read, the smallest allusion among Zendists to the form of the Gathic sentence. The voc, Mazdé standing where it does, and the place — of Ashemché militates against an infin. mriuté(é). Fré vdo ashemché va- ochanhé(é) Mazda Sraé(é)s(k)y@ would look more like an infin. sentence. Srdvayaé(é)ma tim daé(é)nam casts light on Y. 28, 7. Perhaps khshmédvaté had better be rendered ‘your servant’ here in view of sraoté in the next y.
Some might render: ‘I call on You, O Mazda and Asha, to declare what as emanation (@mananhd)(?) of your will is rightly to be comprehended, that I may declare that doctrine, that of your servant, O A’. Pahl. Mrivé(é) was read for mriité(é). Notice vichidydi intelligently rendered as 1* sg, fut. (1 now prefer vijinam). The 15 sg. for 1% pl. is hardly a variation to be noted in such a document, and é@m may be sg. or pl. D. has curiously vddinamam (so elsewhere). Parisphutam = Ashavahishtich (so v. 7, and Y. 51, 4, otherwise = d@shkdrak) arises from the association of A. with the Fire, hence ‘brightness’, ‘manifestation’, or A. may have been misread for ashnd (so). Vazinam, so for vaj°, also in other forms.
7. As to caesura, see vohi apart from mananhd. As sraotd is without expressed subject, some would read sraotd. But khshmdvant in the very possible sense ‘your servant’ may be understood from v. 6, or better the verse belongs after v. 9, see sraoti-- fshénghyo. Notice the three, airyamdn, baé(étu, and verezéna once more. I now prefer Jegibus, ‘laws and institutions’ to donis. Many recoil from a religious sense for frasastim, preferring famam, but the word is associated with yasna and vahma. Some might render: ‘Hear (sraotd) that graciously, O M., hear Thou (sraos tu) (?), OA.; listen Thou, O Ahura, what association for protection, what kith, is it which through its attitude gives good reputation to the circuit?’. Pahl. Read khvéshth. D. om. am in a, but spelt, with DJ. and M.1, nydkshédo; D. has nyokshash (sic vid.) in 6, vddindid in c, min tdno for min at avo, and kardé 7 lak (M1 7 kardé lak) in d. Read khvéshih. Let V. listen, and: = ‘since V.. = amat + pres. Am proves freedom. ‘A listening with V. M. ~ and Ashavahisht’ = ‘VY. M. listens and Ashav°® listens’, instr. sociative. As bae(é)\teus airymanaschd, and verezend = khvéshand, ayarmdndn, variinikich in Y. 46, 1, etc. I have rendered aryamdnth, khvéshih, and varzishno too rigidly here; the abstracts are here free for concrete; read: ‘Who, as client, kins- man-propietor, is a bestower on me, who to the working-class (varzishno concrete (?)) [for thee (°at is gloss)|’, etc. I can only explain sampadayitd as a free rendering. Why the Parsi-pers. renders mashhir is hard to explain. I read shunavad as older than shin®; érmdn?, so in the MS. for yéar°.
8. Fraé(é)stdonhé to fr? (pr?) = ‘most influential (in a friendly sense)’. Others /requentissimi; see fraé(é)stem even applied to ‘evils’ in Vd. 1, 4, but to beneficial influences in Vd. 3, 12 (= ‘most’). Some might render: ‘Grant to F. a steadfast abiding place (?) (in heaven) on account of his piety, and also to me. In Thy Holy kingdom may we be ever the most numerous (?)’. Pahl. The trlr. already thought of the Fire wrvaz, hence his garmak (so); with am understood an orig. bavthtinid (so deciphering) would = 1st pers., so Ner. read, but see the gloss. Knowing well mazbyd, the trlr,
freely renders, as elsewhere (see Y. 28, 7), by mantkané also with free gloss. Sg. 1st for pl. 18 is very frequent, as in hémandnt, see the gl. The text
of the MSS. ato) )9pu is accidental; Y. 48, 3, ¢ we have oj) 24s, SO, cor-