extensive and prolonged study, and this not only consumes a good part of a lifetime, but it leads the toiler through fields of supreme interest. First there is the Veda, and especially the Rik, so different from the oldest part of the Avesta, and so kindred to the Yashts. Here is an
engaging study alone worth years, and to prepare himself for the Avesta
the specialist in Zend should study hundreds of these rich and poetic hymns. Then Pahlavi literature has its gems, as West has so thoroughly shown us, and for an author in Zend the Pahlavi is indispensable. Then the Zend-philolog must see much of the Persian. Surely these particulars alone furnish an excuse for concentration. It is concentration not upon a point but upon a dise, and upon a dise of no narrow dimensions.
As several inteltigent friends and sympathizers have also asked. me what led me at first to Zoroastrianism, I may as well descend still
further into personal particulars and describe an experience of very
great enjoyment as well as profound satisfaction. I had felt an irresistible tendency to interior investigations from early years, and at last began to specialize on the Gnostic philosophy, happening first upon Matter’s
work. Coming to Europe in 1872 I developed this pursuit till it became time to turn to the descendants of Gnosticism in the modern philosophy. But in that rich study I became so fastinated with Kant’s immortal Kritik that I settled upon an exposition of it, and I have my copious MSS. yet with the subject worked up from its foundations and with close detail having availed myself of the advice of the then leading German authorities. I left the Kantian philosophy and its successors for Zend philology to seek the origin of the Gnosis, as I had left the Gnosis for them to search for its results. The, whole series of studies are . closely connected, and each is an independent link in one chain, the history of religious philosophy in its entirety. Surely no one should accuse me of undue concentration after this.
L. H. Mills.
Oxford, Dee., 1890.
Xxui
Additional remarks on the publication of Yasna XXVIII—XXXIV as part I of the work.
On issuing the Ahunavaiti text entirely reprinted I would only add a few words to what was said when I published the Commentary some time ago. First as to the Alphabet: I would retract my remark on p. 447; I no longer think that ) is better than } for m, as }stands in the oldest undated Zend document J.?.
The letter should be curved somewhat even when represented by
j, as it is curved toward thé bottom in J.2 It is however straight in D., and longer than seems natural. Not to be singular I have used 6y in reprinting, but §. stands in J.2, J.8, J*. S(k) 944) stands in J.2, K.5, J.3 in s(k)yaoth, otherwise shy, or sy, appears oftenest, so ashyd appears
in the Vendidiéd MS. given me by Dasttr J. M. Perhaps the additional stroke was the Pahlavi sign for y, as the
character 9% never occurs except as preceding y in the newer Avesta character. In D. and others 0 is used. Except where sh =Sansk. ksh, I have used 3%) = sk uniformly before y in reprinting in order not to differ needlessly from Geldner, and this notwithstanding the usage of the . older MSS.
The distinction between & and w_is carried out in J.?, J.%, but not in J.*, nor in D.: is also used for 6 in my Vendidad MS. & exactly in that form I find in J.* as far as to Y. 44. 2, 0 where a different writer begins who uses CG. In D. it occurs in 8(k)yaothna.
J.?, J. have ik more erect than Go 32 has both GC, and lL apparently,
and in somewhat close proximity, but the more erect form turns to the
left at the top. I have only noticed Cin the Vendidad MS. My printer
did not appear to have a type which could represent l, erect when
printing the texts.
I must also mention that I have collated still one more MS. of the Pahlavi. It was from Haug’s Munich collection no. 7 under 6a, 6 which contains scattered fragments of the Gathas with Pahlavi translations, but as they seem very much more extended than the others, it was hardly
fair to publish the differences as variations of the same document. In reprinting I have given a partially restored text in the transliteration. Schleicher first suggested the most frequent of these reconstructions, Spiegel following with many others, but Roth through his able pupil Aurel Mayr applied the principles which had been established to the first four Gathas; see the interesting treatise Resultate der silben-zih- lung aus den ersten vier Gathas (1871). This was well reproduced and extended by Bartholomae in his valuable edition in 1879. I regret that I omitted applying this improvement in the parts longest printed, as it is very useful although not everywhere producable without fault.
The letter ere = Sanskt. 77 should be pronounced as one syllable, 80 also one syllable only should be sounded wherever a curve is under two vowels, 0, ete. Hvé sometimes = huvd as similar Vedic words are
often to be pronounced, the false writing jyd Sonldd is explained as for jivd® Ca rap ), etc., ete.
With regard to exegesis, I would remind beginners that all exten- ded works are apt to be more conservative, rash suggestions being large- ly confined now-a-days to short articles in periodicals. I accordingly endeavour to present my more hazardous proposals in alternative ren- derings, not thinking it desirable to fasten uncertain opinions upon in- experienced readers merely for the sake of an affirmative style, a mode of procedure better adapted to a science later on than to one the ma- terials of which have only now been fully examined.
I need hardly remind Zendists that I endeavour at every step to improve on my renderings in the Sacred Books of the East, XXXI, (1887). I trust that I have not wantonly varied from them, but five years is a considerable interval, and most Zendists seek to vary their views at every publication. I have not made myself an exception, for I have even endeavoured to improve in different parts of this same book. The Commentary was printed while the texts were in manuscript, and in printing the texts later I have suggested improvements up to the last, and I will continue to do so. I may even adopt the proper letter v where I have elsewhere followed the old custom of using the German letter w for it; this to avoid misleading beginners.
As to the Pahlavi text, an estimable Parsi friend once told me that he thought it would have been better had I produced the Pahlavi in its
original character without transliteration , affording only a translation, and leaving scholars to transliterate for themselves. But I need hardly say that one third of the labour consists in the decipherment, which is at times so difficult as to baffle us entirely. It would have been an easy device to send the copied characters to the printer, With regard to the translation of both the Pahlavi and Neryosangh, but especially with regard to the former, I would repeat with greater emphasis what I said perhaps too moderately on pp. XIV, XV, XVI, XVII. A treatment of these commentaries without the closest regard to their original is in my judgment totally unscientific, and can only lead to spurious results. The translation of the Pahlavi especially should be broken up by a con- stant effort to follow its original, and exposition should intervene at every call for it. That sprightly vivacity which we so value in the translation of original matter (which the Pahlavi in these translations is not) would only destroy what likeness the translation bears to its original, and so give beginners an excuse for neglect.
As to Neryosangh I would only add that he should not be read in the light of hitherto quotable Sanskrit, for that would be to blunder at every step; but he should be regarded as himself affording usage quotable in future Sanskrit lexicography.
Beyond all doubt he uses words in a sense correct as attested by his contexts and his originals, but in a sense discarded as unquotable from the later dictionaries.
And he by no means invents these definitions; they were once pre- valently in use when he was taught Sanskrit, as we see from the older dictionaries, and the quasi artificial character of all the later Sanskrit should make us cautious how we condemn meanings given to words even by such an irregular writer merely because they have not been found applied in the same way in more classical productions.
I would especially recommend Burnouf’s Sanskrit dictionary to be used constantly together with the other lexicographical works, but the remoter date of Burnouf’s book should of course be borne in mind, and it should be used with every caution.
As to the former practice of citing Neryosangh untranslated and as if he presented a rendering of our present Pahlavi texts see above on p. XVI,
We may say of his work in its entirety and with all its imperfec-
tions that when it, is understood, as it only can be by the most careful study of its originals, it turns out to be, or at least to have once been, one of the most important texts in the language in which it stands as well as one of the most difficult, for it has assisted in giving us that original exegesis of the GAthas which first taught us where to begin, and its usefulness continues,
The Parsi-persian however vies with, or even surpasses Neryosangh in value because it gives (or “gave’) us the most direct clue to the at times almost inserutable Pahlavi, and it must be itself a descendant of Parsi translations of the Pahlavi which existed from times long previous to Neryosangh and shortly after the Arabic had pervaded the purer Persian of the priests.
With regard to my free rhythmical reproductions I would ask those to whom English is not vernacular to remember that the accent in English, as it does in German, alone brings out the metre, and unless the English accent is familiar the rhythm will be wholly lost. This re- mark applies also to the entire volume in the Sacred Books of the East, but especially to Y, IX—XI which ought to have been printed in the form of poetical verses, many syllables being freely thrown in to improve the harmony.
One important word to scholars interested in the science of Com- parative Religion. One scholarly gentleman cited by Professor Cheyne (see his Bampton Lectures for 1891, p. 434) was so far Scandalized at the uncertainties of the Gath4s that he feared, as he said, ‘even after Mills’ translation’ to use them as materials in Comparative Theology; but it should be well noted that the uncertainties of the Gathas chiefly con- cern closer detail. The terms which they use are such, and the cha- racter of their syntax is such that they almost always afford us alter- native certainty. If one idea is not exactly expressed, then another closely kindred to it is present.
To the philolog the difference is great, and the controversies will probably only be decided by external interests, but to the student of Comparative Theology it ig very often really indifferent what one of two, three, or even four somewhat differing casts may be given to any single strophe, as each view abounds in the characteristic religious tone of the whole, and where a particularly striking idea may be only pos- sibly present in one place it may generally be found without any doubt
Xxvit
in some other!. The uncertainties of the Gathas should trouble Compa- rative Theology scarcely more than those of the Old Testament, and this can be easily verified from this or other books.
There should be some profit for specialists in the general grouping of the differing translations aside from their more immediate study.
The differences in these historical renderings in the Pahlavi, San- skrit, and Persian from those adopted by us of to-day should be highly salutary in the lesson which they convey. The imperfections, self- contradictions, and at times puerile suggestions which appear in them furnish one of the most interesting cases of traditional uncertainty ever succinctly presented, and are nearly as important in warning us against implicit confidence in ‘tradition’ as their, on the whole, astonishing
- approaches to accuracy are important to warn us against its neglect.
Specialists in Vedic and even those in Semitic exegesis should read these commentaries carefully if only to learn how curiously ancient critics could err, as well as how wonderfully they could succeed; and they should apply the lesson gained to their own departments, neglecting no ancient hints, but taking special care not to follow them implicitly, above all things to distrust what seem to be at first sight their indications as to grammar.