Thesaurus Indogermanischer Text- und Sprachmaterialien

The Avestan Manuscript J2
Introduction by L.H. Mills





Reproduced in Facsimile











A.D. 1323



New York

Only two hundred copies of this MS. have been reproduced;
of which this is No. 78.


THE accompanying pages contain a reproduction of an ancient-possibly the most ancient-dated MS. of the YASNA with a commentary in Pahlavi. This MS. has been presented to the University of Oxford by Dastur JAMASPJI MINOCHEHERJI JAMASP ASANA, a High-priest of the Parsees in Bombay, on whom the University of Tübingen has conferred the degree of Doctor of Philosophy and the University of Oxford that of Honorary Doctor of Civil Law, in recognition of the valuable services rendered by him to the study of Zend literature. It originally came into his possession by inheritance.

The following is the Dastur's history of his gift. derived from a letter of December 3, 1892. The MS. was written at Cambay in the year 692 A.V. (1323 A. n.) by one Ervad Mehrban Kai Khusru. One of the ancestors of the donor, Jamasp Asa, a well- known Daslur of Naosari, obtained it at Broach, whither he had gone to study Oriental languages. After Jamasp Asa's death, his Pahlavi and other MSS. passed to his third son Dastur Behram. as his share of his father's library, the contents being equally divided among Jamasp Asa's four sons. On the death of Dastur Behram this MS. passed into the hands of his sons and grandsons, Dastur Minochehr. Tehmur, and Jamshed. All these Dasturs were celebrated for their learning, and especially treasured this precious MS. After the death of Da-slur Jamshed. however, its importance appears to have been lost sight of until thirty-seven years ago, when the donor found it among a heap of other books in the house of his father-in-law at Naosari. It was then much damaged, but has since been most carefully preserved.

It was lent by the Dastur to a few scholars in India and Germany, as well as to myself here, so that its value became recognized, and some tempting offers were made for its purchase. The Dastur, however, courteously declined to sell this unique MS.; but in order to secure its safe keeping for the future, he, at my suggestion, generously offered it as a present to the University of Oxford, for deposit in the Bodleian Library. The only condition which the High-priest attached to his gift, was that the MS. should be once more returned to Bombay for the purpose of being copied before its final departure from the East. In order to avoid the risk of sending the MS. back to Bombay-for it had already suffered considerably in its transmission to Tübingen and Oxford-the University decided to incur the expense of having the whole collotyped at the University Press, sending copies to the High Priest at Bombay, and publishing an edition of 300 copies for the benefit of Oriental scholars in Europe.

The YASNA consists of several sacred compositions of widely varying age and character, grouped around the GATHAS, the ancient Hymns of Zoroaster (Zarathushtra), the Founder of the original Parsee religion.

These HYMNS are composed in various metres, similar to those of certain hymns of the Rig-Veda, although no trace of the metrical arrangement is visible in this or any other Zend MS. The discovery of their metrical character may have been earlier made by another scholar, but Professor Westergaard in his edition (1852-54) arranged them in metrical lines. Professor Spiegel adopted the same treatment in Ills edition, published in 1858.

These Gathas are the oldest, as well as the most important parts of the YASNA and of the entire AVESTA, and their age has been variously estimated at from 3400 to 2600 years. They may be older than the first age mentioned, but they cannot well be later than the last. They are found in our MS. between folio 174, page 348, seventh line, and folio 221, page 442, second line; folio 236, page 472, thirteenth line, and folio 294. page 587, eighth line ; folio 295, page 590, sixth line, and folio 301, page 601, ninth line.

The next oldest portion of this collection is that known as the YASNA HAPTANGHAITI, that is to say, [p. VI] the Yasna of the Seven Chapters. It describes forms of sacrifice to Ahura Mazdah and to his personified attributes, the Amesha Spenta (Amashaspends), that is, to the `Bountiful' or `Holy' Immortals, to the Fravashis or Guardian Spirits of the Dead, to the Sacred Waters, to the Soul of the Herds, ending with various shorter prayers. It occupies the space from folio 221, page 442, line six, to folio 236, page 472, line nine.

After this in order of age follows the SRÔSH YASHT, which is a devotional address in metre to the personified Obedience, Sraosha. It begins on folio 307, page 614, line six, and extends to folio 321, page 641, line six. Next in antiquity we may place the HAOMA YASHT (Hôm-Yasht), which corresponds to a certain extent both in thought and expression with some of the Soma hymns of the Rig-Veda. It is a collection of metrical pieces beginning at folio 82, page 163, line eight, and ending at folio 107, page 213, line ten. Chapter XII may be regarded as a Confession of Faith. It extends from folio 114, page 228, line three, to folio 119, page 237, line twelve.

Metrical matter also occurs here and there throughout the rest of the Yasna.

Chapters XIX, XX, and XXI contain commentaries on the formulas Ahuna-vairya, Ashem Vohu and Yeñhê hâtãm, which are brief but especially sacred prayers. They extend from folio 136, page 272, line one, to folio 151, page 301, line one. These sections need no special mention, as they have all been described and translated by me in volume XXXI of the Sacred Books of the East published by the Clarendon Press.

The MS. gives the entire1 Zend text of the YASNA, accompanied by the ancient Pahlavi translation and commentary, sentence by sentence, although certain sentences are wrongly divided both in the Text and in the Commentary. So far as is known at present there are still some thirty-three MSS. of the YASNA extant and accessible, but only eight of these possess the Pahlavi translation. There are others containing the Zend text of the YASNA interspersed with that of other books, and there is one manuscript known containing the Pahlavi translation only, without the Zend text.

Our MS. was completed -- according to its colophon on folio 385, page 770, written by the same hand as that on other folios -- by Aêrpat Mihr Âpân Kai Khûsrô (or more fully Aêthrapaiti Mitrô Âpânô Kai Khûsrôbô), the transcriber of three other venerable Zend MSS., on the second day of the first month in the year of Yazdakard six hundred and ninety-two, that is to say, the twenty-sixth day of January A.D. 1323. This date differs from that commonly given heretofore, and accepted by the present writer on page xvi of the Preface to volume XXXI of the Sacred Books of the East, where a period is referred to which corresponds to the nineteenth day of the eleventh month of the same year.

A serious difficulty has always been felt in fixing the relative age of this codex and that of the famous MS. in the University Library of Copenhagen, No. 5. It was generally supposed that our MS. was completed only twenty-two days after the completion of the other that is, between the twenty-seventh day of the tenth month A.Y. 692 (November 17, 1323) and the nineteenth day of the eleventh month (December 9). This would involve the supposition that Mihr Âpân Kai Khûsrô worked on the two MSS. at the same time. But it is not very probable that he did this, or that he hurried over this document of 770 large pages within such a limited period as three weeks. We therefore look at the colophon more closely, and we find that the handwriting degenerates considerably, as if the writer had become weary. Six errors at least are corrected by Mihr Âpân himself on this one page, while two remain uncorrected. Among these we notice the word yôm inserted over the line between the words fravardîn and shnat. This hasty though genuine correction (it is from the original hand) we have hitherto taken as it stands, reading vohûmanŏ birakh fravardîn yôm shnat-î DCXCII, yazdakardîg: `on the vohûman month and the fravardîn day,' that is to say, `on the nineteenth day of the eleventh month, A.Y. 693.'

But in view of the difficulties mentioned above we are obliged to question this emendation, or still further to emend it. And the only manner in which this can be done is to suppose that this superlinear correction was carelessly inserted in a wrong place by the copyist, after he had finished his protracted task.

The first yôm (day) at the beginning of the colophon may have been set down in mere absence of mind, as the writer's attention was for the moment fixed upon the words which were to express the actual date, and not upon the words yôm and bîrakh. When he had finished the first line in his usual handwriting, his eye seems to have caught the misplaced yôm above, and he did not repeat it in its proper place, which was perpendicularly beneath the position which it now occupies. He may have intended to correct it, but [p. VII] omitted it, till he finally took up the paper and hastily put down the objectionable word as a superlinear correction where we now see it.

There is also a linguistic argument against the application of vohûman to bîrakh as an adjective with the meaning of the `vohûman month,' as also against the `fravardîn day.' The adjective before the noun would not be incorrect, but it would not be so usual a form; whereas if an accidentally omitted yôm were replaced before vohûman the sentence would fall into a natural order, the noun preceding the adjective. It would then read, `on the day vohûman, and the month fravardîn.'

We should also remember that the MS. was a piece of work for which the copyist was paid, and that he may have been reluctant to disfigure the appearance of the page by erasing the most prominent word on it. His employer would naturally care little for the accuracy of a date which only concerned the scribe whom he had employed, but he might have been displeased at an unsightly emendation. A similar reluctance to correct mistakes appears in the work of the more careful copyist of the Zend-Sanskrit Yasna, the third in Dastur Jamaspji Minocheherji's catalogue, and now, by the generous gift of its former owner, likewise in the possession of the Bodleian Library. The writer merely places short perpendicular strokes over the superfluous or misplaced characters, and this evidently arose from a reluctance to mar the appearance of the sheet.

Such are our explanations of the imperfect colophon of this MS. They are the best that can be offered to meet the formidable difficulties in the way of the hitherto accepted opinion, and they lead us to believe that our codex was completed not twenty-two days after November 17, A.D. 1323, when the twin MS. now in the University Library of Copenhagen was finished, but nine months and twenty-five days before its transcriber had penned the last stroke on that justly celebrated document2.

The mechanical execution of this MS. is satisfactory, although it shows traces of haste. Its pages of fifteen tines each (not of course always complete) contain, when full, on an average ninety-one words, of which twenty-four are Zend and sixty-seven are Pahlavi, the Pahlavi, through the absence of the short vowels, occupying perhaps one-third of the space taken up by the Zend with its complete signs for every sound. From the freedom and flow of the handwriting the words, especially those in the Pahlavi language, need a practised eye to read them, as they are very different from those in more deliberately written codices, and still more from those of our printed texts. The folios are paged throughout on one side with Sanskrit, on the other with Pahlavi or Arabic figures. There are a few mistakes. The Pahlavi pagination, however, continues correct throughout; except where it is lost through the margin of the page having been injured. Page 239 is numbered 238, the numbers running on till page 311, when the Sanskrit pagination becomes again correct. After 333a the Sanskrit numbers are one in advance; after 362a they are four in advance, and so continue to the end. Folios 1, 2, and 150 (pages 1-4, 299, 300) have long been missing, and the paper has suffered in many places from the effects of moisture and from insects. For the most part these lacunae are of moderate dimensions, and they can easily be filled up by a reference to other MSS. and to the printed editions. More serious injury has happened between folios 113 and 152; while on folios 212, 213, 214, 217, 2i8, 219, 220, the damage has been very great. About one-half of each of these folios has disappeared, or has fallen into fragments too minute for reproduction. These fragments have been carefully preserved between plates of glass, and may be consulted at the Bodleian Library.

The service rendered by the present publication to Zoroastrian science is very great. Familiarity with MSS. is essential to a critical study of Zend literature, and scholars in every part of the world will now be able to consult this treasure in their own homes.

The Pahlavi Translations in the eight extant codices which preserve them are of special importance. Though the MSS. from which the existing codices were copied cannot have been prior to the Sassanian period, their real originals must have been much older, for there must have been efforts to explain the Avesta from the first, and it is difficult to believe that they left no results. They possess also a value as evidence to the existence of lost texts, and written texts of the Avesta must have been attempted at a very early period. In their later forms these translations were of course but rarely original. But the MSS. which we now possess were for the most part more or less careful reproductions of previous codices, with some few original or selected variations. As stated above, the Pahlavi Translations generally follow the Zend texts word for word; but the copyist, unless himself a critical scholar (which was not often the case), in selecting [p. VIII] emendations of his Pahlavi texts might, and probably very often did in his embarrassment, adopt Pahlavi readings from other codices without noticing that they translated a Zend text differing slightly from that which he had just penned, so that his modified Pahlavi translation, though failing to translate the exact forms which precede it, may render an equally valuable service; that is to say, the hasty renderings in Pahlavi, where they fail to reproduce their Zend original, as written beside them, may shew us Zend readings now lost to us. So long as the copyist made excerpts and avoided originality, his very faults may become valuable to us. On the other hand, some of the supposed inaccuracies of the Pahlavi Translations are useful, because they may be suggestions of texts conceived to be emended by those translators themselves rather than helpless fatuities. The translator at times could not accept the Zend text which he saw before him, and therefore changed it, in order that what seemed obscure to him might become intelligible. Some of these hazardous emendations seem to have been successful, for they at times suggest to us better readings. How far these curious attempts prevail through the Pahlavi text of this MS it will not be safe to estimate until the whole of it has been edited as deciphered and translated. Those portions of it which render the GÂTHAS have been elaborately so treated and commented upon by the present writer, in a work which formed the preliminary study to the translation of the GÂTHAS in the Sacred Books of the East.

How far Mihr Âpân Kai Khûsrô exercised any original scholarship in the present production, it is difficult to Judge. We may, however, notice that he omits to name any original from which this MS. was copied; whereas in the case of his other work, the Zend Pahlavi Yasna No. 5 in the University Library of Copenhagen, he does not fail to state what its original was. The present codex may therefore be considered in some respects as an edition (if such an expression be allowable) rather than a copy. In either case this contribution to the study of his native scripture was of the greatest value in his day, and, as centuries have passed, its remote evidence has become most precious. It is to be hoped that its publication will stimulate the pursuit of the studies for which if forms so important a document. Zend philology is a branch of Oriental research as wide as any other, and in some practical aspects it may even be said to have claims superior to those of most of its kindred subjects. It not only includes a wide range of materials for linguistic inquiries, but it enables us to observe the origin and growth of some of the most important religious and philosophical doctrines. This, its oldest complete codex, often expresses lofty sentiments; its most important parts have some claim to poetical grandeur, and it possesses a technical interest also as one of the earliest specimens if not the earliest of metrical composition. Zend lore as a whole is rich in material illustrative of the early history of mankind.

May, 1893.


1 Save where small portions of it have perished, as described below. back

2 See a paper read at the Ninth International Congress of Orientalists in London before the Persian Section on September 10, 1893, by the present writer, where a somewhat different view is taken of the first occurring word yôm. back

Copyright of this digitised version Jost Gippert, Frankfurt 2001-2002. No parts of this document may be republished in any form without prior permission by the copyright holder. 15.4.2002.