[In the essay below, Waterhouse examines the process by which the Avesta, including Zoroaster's Gathas, was compiled and discusses the structure of each.]

The discovery of the key to the understanding of the Avesta, the Bible of the Parsis, is a romantic story. In the year 1754, a young Frenchman, Anquetil du Perron, saw a few pages of a manuscript in an unknown Oriental tongue, in a library at Paris. His interest and curiosity were so awakened that he determined at all costs to decipher the writing. He thereupon joined the French East India Company as a ranker, and embarked on a ship bound for Bombay, the centre of the Parsi community in India. After a hazardous journey, he reached his destination, and the French authorities honoured his purpose by releasing him from duty, and by granting him a certain amount of support. Although at first du Perron could find no one to teach him the language, eventually, through bribing an erudite priest, or Dastur, he acquired the requisite knowledge for his task, and also secured nearly two hundred manuscripts. The Parsis were very suspicious of the intruder, and he had to return to France to do his work of translation. This occupied du Perron ten years, but in the year 1771 he was able to publish a volume which he called Zend-Avesta, Ouvrage de Zoroastre. That this title was misleading will be shown later, but du Perron had rendered magnificent pioneer service to Iranian scholarship. The reception given in Europe, however, to his work was by no means favourable, as the Persian scholars of the day, such as Sir William Jones, did not believe in a separate Avestan dialect, which was postulated by du Perron.

Although the discoverer was right in his assumption, the translation he then made cannot now be relied upon, as it contains many inaccuracies and reflects in parts his somewhat credulous disposition. Nevertheless, it was to be expected that du Perron's work would be inaccurate as the Parsis themselves knew the translation of their own Books only through the medium of a disused language, Pahlavi. The verbal traditions, however, that du Perron had received from his Parsi instructor were of great reliability. Many of the names of ancient Persian deities, hitherto unknown, were revealed, and most valuable light was thrown on old customs and ceremonial. But the authorities decried du Perron's work as a hoax, and naturally it was pointed out that the contents of the Avesta to a large extent did not accord with what was known of the teaching of Zoroaster. The lack of resemblance between the Avestan language and modern Persian was held to be sufficient reason for denying the genuineness of what du Perron had produced. Although du Perron's work gained some credence in France, the general opinion of European scholarship was hostile to it.

Some fifty years later, another Frenchman, Eugene Burnouf, who believed in the so-called Avestan tongue, carried out further research into the philology of the language, and corrected many of du Perron's obvious mistakes. Burnouf did valuable work on the grammatical side, but did little to sift the manuscripts before him as to relative importance in matter of date or doctrine. It was, moreover, not yet realized that the Gathas were the only likely relics of the writings of Zoroaster himself. But Burnouf had made possible the work of many subsequent investigators such as Bopp, the grammarian, Brockhaus, the lexicographer, and Westergaard, who published a monumental compendium of the Parsi religion and literature. These and many others have presented to us through their researches both in Sanskrit and Pahlavi, a fairly reliable text of the Avesta. Perhaps the most comprehensive translation yet undertaken, however, is that by James Darmesteter and L. H. Mills, in the Sacred Books of the East. But this translation cannot be taken as final, as much has been added to the knowledge of the language since the work was undertaken. Of course, previous to the discoveries of du Perron, Europe had not been without Persian Books of great antiquity, but few could be understood. The statements of Greek and Roman writers concerning Zoroastrianism had been collated, and Thomas Hyde, a famous Oxford scholar of the early eighteenth century, had published a ‘magnum opus’ on the subject,¹ but it did not show much knowledge of the original sources.

Although the expression Zend Avesta is widely found, and was used originally by du Perron, it is not an accurate description of the Bible of the Persians, as the term means ‘commentary on the Avesta.’ The derivation of the term ‘Avesta’ is uncertain. It is possibly akin to ‘Veda,’ i.e. Knowledge, ‘vista’ being the past participle of the Sanskrit root ‘vid,’ to know. More generally it is taken as coming from an ancient Avestan form ‘upasta’ which means ‘the original text,’ or ‘scriptures.’ Since Zend means ‘commentary,’ Avesta, simply, is the better title. As we now have it, the Avesta is only a fragment—though substantial at that—of a much greater literature, a great deal of which, some say two thirds, has been either lost or destroyed. A very small amount belongs to the period of the Prophet himself, though Pliny attributed to him two million verses. Zoroaster's output may well have been considerably greater than the Gathas: such an estimate as Pliny's, though of course absurd, affords, however, some indication of the immense labours of the followers of the Prophet.

The contents of the Avesta represent a long period of diverse development, and there are few religious Books in the world which present so many literary problems to the investigator. The original Avesta is said to have consisted of twenty one Nasks, or volumes, one for each word of the Ahuna Vairya formula (See ch. 1). These were divided into three sections, each of seven groups, containing an encyclopaedic account of Zoroastrian history, devotion and science. The ‘science’ was chiefly of an astronomical and astrological character. The Avesta was at first jealously guarded, and copies were preserved in the chief Fire Temples. The ravages of Alexander the Great (b.c. 330), however, were responsible for the loss and dislocation of a large number of Persian manuscripts, the nature of which can now be ascertained only through the evidence of those Books, such as the Dinkart, which contains summaries and quotations from the missing material. Diodorus the historian records that when Alexander burnt Persepolis, he put to death some of the leading scholars who lived there, and their manuscripts perished with them, as the Achaemenian stronghold was the repository of many Persian archives. The Arsacid king Valkhash (c. second century b.c.)² was probably the first to inaugurate a collation of the various scattered manuscripts, and also caused many oral traditions of Zoroastrianism to be committed to writing. But it is doubtful whether the priests of those times could compose in the same language as the original Avesta. It was not until Sassanian days (third and fourth centuries a.d.) that a comprehensive collection of Books was made. The first Sassanian monarch, Ardashir-i-Papakan, (226-240), by the help of his high priest, Tansar, produced a text of the Pahlavi writings. Shahpur I (a.d. 241-272) and Shahpur II (a.d. 309-380) were instrumental in gathering further texts from different parts of India and the Roman Empire. Adarbad Maraspand, the prime minister of Shahpur II, made an arrangement of all the resultant discoveries, and a Canon of Scripture was announced.

This Canon of Parsi scripture includes all that survives in an extinct language, Avestan, which few Oriental scholars can understand. In fact, very few Parsis know it, and many of the Zoroastrian priests of the present day are compelled to recite their sacred Books in words utterly meaningless both to them and to their listeners. The modern translations of the Avesta, accepted by the Parsis, are in part most inaccurate, and certain stereotyped and traditional interpretations are now accepted without question. The later Pahlavi writings have, for instance, seriously changed the names of many of the old religious terms of Persia, and the Pahlavi translations of the Avesta do scant justice to the greatness of the finer passages, especially in the Gathas. One result has been that Darmesteter's translation of the Gathas, founded largely on the Pahlavi, is rendered practically useless, as the translator did not rely on the Gathic Avestan, the only true text. Incidentally, Darmesteter relegated the whole Avesta to the period of the Neo-Platonic literature, as he considered the religious allegories of the Gathas to resemble those of Philo. His argument was that, before the Greeks, there had been practically no philosophical development—a very hazardous hypothesis. Carnoy has aptly suggested that Darmesteter's mistake was due to his identifying ethical abstractions with philosophical concepts. That many of the writings are the product of activity in the Christian era will not be disputed, but the Gathas, Yashts and Vendidad are among the documents of far earlier origin. It is extremely difficult to assign dates to many parts of the Avesta, and sometimes the only reliable evidence that can be so used is internal. The style and metre of the contents are rough indication of the period to which they belong. Many of the texts, however, are corrupt. An exception to this is the text of the Gathas. The conclusion is that the Gathas were regarded as especially sacred, and have suffered little from editors. This strengthens our hypothesis that they represent the actual teaching of the great Prophet. In the later Books textual alterations abound, together with obvious blunders and intentional scribal glosses.

We have in the Avesta a library of Books which extends, like our own Bible, over a period of a thousand years. But revelation, in the Avesta, does not culminate at the end of the period as it does in our own Bible: generally speaking, the finest Avestan works are the earliest chronologically. The process of compilation of the Avesta was roughly as follows. The sayings of the Prophet and those who immediately followed him were the first to be recorded, and are found in a dialect called ‘Gathic Avestan.’ These records were then edited and elaborated by successive generations, who used practically the same language. In process of time there was necessitated a new explanation of the original sayings. This was supplied in a similar dialect, known to us as ‘Younger Avestan,’ and was called Zend. After another lapse of time, the old sayings and the commentary on them were looked upon as one, and equally sacred. But since the various forms of the Avestan language had become obsolete, another Zend was made on the extant scriptures, in Pahlavi, which was the ordinary tongue of Persia during the Sassanian era.

Finally, there was the Pazand, or re-exposition of the whole lot. When Zend is spoken of as a language to-day, Pahlavi is usually meant, because the Zend language, if ever it existed as a separate tongue, has dropped out of use. Originally the term Avesta was used only of the writings of the Prophet and those who were closely associated with the early propagation of the faith. Later on, a wider significance was adopted, to include the commentaries, and because of the increasing difficulty of sifting the wheat from the chaff, all the traditional writings became adorned with the same halo of sanctity. It was not generally known what was from Zoroaster and what was not, and there was no doctrine of degrees of inspiration. The ancient Greek and Roman writers, as well as the Persians themselves, ascribed a great deal that was obviously not Zoroaster's to his authorship. Just in the same way, Moses was long accredited with the account of his own death! The difficulties involved in the theory that Zoroaster was responsible for the bulk of the Avesta were overcome by the assumption that the Nasks were dictated to the Prophet, as they stand, through divine revelation. Thus chronological obstacles were neatly swept away, as God could dictate words and phrases appropriate to a generation to come, and was not limited to the vocabulary or thought-forms of Zoroaster.

There is no statement in the earliest Persian Books as to who transcribed the narrative of the experiences or message of Zoroaster. Unlike Muhammad, the Prophet of Iran does not seem to have been commanded to publish the visions he had received. It is possible that Zoroaster himself wrote nothing, but that his friends and disciples transcribed his sayings; parallels to this may be found in the composition of the Christian Gospels and the Vedic hymns. But there is an atmosphere of personal reminiscence in the Gathas which would suggest that the Prophet at least dictated some of the Yasnas. The oral tradition which is manifest in many portions of the Avesta is responsible for a certain amount of inaccuracy. Haug thinks that the majority of the Parsi scriptures were written down when their language was unintelligible to the scribes. Various additions to the manuscripts were made by the high priests in particular. In later days it was believed that the high priests, who claimed to be in the succession of Zoroaster, were given inspiration to interpret his mind without fail, and had authority to add to his writings. This may partly account for the tradition that Zoroaster was responsible for the whole of the Avesta.

The languages and dialects of the Avesta, which have been outlined, are all inter-related. The Avestan, in which the majority of the Books is written, is a very rich tongue, making especially effective usage of compound words which add colour and charm to the narrative. It belongs to the Aryan family of languages, and consequently has affinities with Greek, Latin, Sanskrit and Teutonic forms. Of these, it is perhaps closest to Sanskrit, as it is merely a different dialect of the Sanskrit in which the Vedic books are found. Haug maintains that the two dialects are as close as, for example, the Ionic and Doric dialects of Greek: ‘As the Ionians, Dorians, Aetolians, &c., were different tribes of the Greek nation, whose general name was Hellenes, so the ancient Brahmans and Parsis were two tribes of the nation which is called Aryas both in the Veda and Zend-Avesta; the former may be compared with the Ionians, and the latter with the Dorians. … There can be no doubt that classical Sanskrit was formed long after the separation of the Iranians from the Hindus.’³

As to the relationship between the two main Avestan dialects there is considerable dispute. It is difficult even to prove that Pahlavi is less ancient than the Gathic, as the latter may just have been a tongue spoken in a different part of Iran. But grammatically, the Gathic is more primitive, and the bulk of evidence favours its greater antiquity. Unfortunately, we have not much knowledge of the grammar of either dialect, and since the original forms of many words are now beyond recall, any translation is bound to be inadequate. Because of this, we have to pardon the great variations in the rendering of the Avesta, which are at times at least disconcerting. When the translations of Moulton, Darmesteter, Haug and Spiegel, for instance, are compared, it is sometimes quite a task to recognize the same passage so variously translated! The available texts have, moreover, presented the greater difficulty to the translators in as much as many of the Parsis care little for the sense of the words in their scriptures, and are content if they are able to recite them.

In the Pahlavic sections of the Avesta, translation has been found easier, especially as there is a modern form of Pahlavi spoken to-day. The ancient form contained a number of Semitic words which had been incorporated during the period when Israel had close contacts with Persia; to which period reference will be made later. Modern Pahlavi, however, has assimilated Arabic forms to the general exclusion of the Semitic. A great need is for modern Parsi scholars to produce a critical edition of their own scriptures; too long have they relied on the researches of Europeans and Americans. It is reported that the Dasturs of Bombay are attempting this task, and it is to be hoped that their labours will bear quick fruition. Since the Great War, and the untimely death of James Hope Moulton, there has been very little work attempted in the realm of Iranian studies.

It now remains to describe and discuss briefly the contents of the extant Avesta, with some reference to the so-called ‘deutero-canonical’ literature. The Avesta proper contains the Yasna and Gathas, the Yashts, the Visperad, the Vendidad and a number of smaller Books and fragments. The Gathas are the most important part of the Yasna. They have been described as ‘metrical sermons’ and contain most of what is known of the teaching of Zoroaster himself. Their metre is not of a nature that lends itself to translation in English verse, and it is unfortunate that the remoteness of many of the Gathic allusions and the inadequacy of any rendering of them have put these remarkable poems out of the reach of the general reader, together with the rest of Persian poetry. There has, alas, been only one Persian poem which has made special appeal to the West—the Rubáiyát of Omar Khayyám. But this was largely due to the wonderful translation of Edward Fitz Gerald. Says Moulton: ‘Would Omar have been heard of in English literary circles had (say) Carey translated him?’⁴ But Omar comes from a period very different from that which is our present concern, for the date of the Gathas is something like three thousand years ago. Considering this vast age, it is more than remarkable that such fine literature, nobler by far than the later Avesta, should find birth in a primitive civilization. There is more colour and movement in the language of the Gathas than in the somewhat conventional phraseology of the subsequent Books, and the strong personality of the Prophet outstands through all these poems that tell the story of his ministry. Although their language is obscure, and the text mutilated, there remains, as Söderblom remarks, ‘a notable sureness of style. Comparison with the Veda hymns of the earlier period leads us to suppose that the art of verse was highly developed even in Indo-Iranian times before the eastward bound Aryans had been split between India and Iran. Zarathushtra or his friends were able to move in this heavy, artificial armour, so to speak, and these queer, twisted, complicated, and abstruse expressions. From these complicated verses and stanzas we can picture a very early and simple civilization, and a burning prophetic zeal.’⁵ We need to remind ourselves that a simple civilization often may produce great poetry. Examples of this abound in the Old Testament, whilst Arab farmers and shepherds have shown no small poetic gifts.

The metre of the Gathas is very different from that of the later Avesta: Moulton likens it to that of Hiawatha, as it is also octosyllabic. One Gatha, however, is in prose. This is the Yasna haptanghaiti, or Seven Chapter Gatha, a collection of prayers addressed not only to Ahura, but to his surrounding Spirits (the Amesha Spentas) and the genii of the elements. In style and in doctrine this Gatha differs greatly from the others; it is later, and represents the trend of the faith after the death of the Prophet. The old Nature deities which Zoroaster had banished, here return, though Ahura Mazdah is still supreme. A parallel process which may be observed in the early history of Christianity: some of the dethroned pagan gods returned as Christian saints, with scarcely a change of name. Later still, when Zoroastrianism came under Magian influence, yet more banished deities were restored.

In all, the Gathas are seventeen hymns, or poems, and their arrangement is according to their metres, of which there are five. The first word in each section determines their name. The Ahuna Vairya formula, which has previously been referred to, is also appended to the Gathas, together with a few other passages especially suitable for repetition, such as the Ashem Vohu, a proverbial saying to the effect that Right rights itself.

The Yasna, of which the Gathas are a part, has seventy two chapters, and is a liturgical compendium meant to be used in connexion with the ceremonies connected with the drinking of the Haoma plant. There are three divisions in this Book: invocations, hymns and commentary. Some of these hymns are known as Yashts, and are dedicated to Angels. But there is also a separate Book of Yashts, comprised of twenty one poems of praise. Chief among the Angels, or Yazatas, are Mithra, Anahita and Sraosha. Some of these Yashts are very fine literature, but they are most unequal in merit. It is likely that many belong to the Achaemenian period, though they incorporate and adapt earlier material for purposes of worship under later conditions. The Visperad is included in the Yasna, and is a series of disconnected invocations to many deities—thus by derivation its name means ‘to all the lords.’ It personifies many of the heavenly bodies. The Vendidad means ‘law against the demons,’ and is a late Book dating from about the second century B.C. Many early legends are recorded in this work, such as the story of Yima, the Persian Noah. The story is also told of the temptation of Zoroaster to worship the Daevas. Three chapters are devoted to instructions as to the way in which true Zoroastrians treat their dogs—as dogs are regarded as sacred creatures among the Parsis. There is also a list given of things clean and unclean, as touching ceremonial, and methods of avoiding defilement are specified in detail. This strange Book is mainly cast in dialogue form, the Prophet asking questions and receiving answers of Ahura. The above works, with the addition of various small fragments, complete the contents of the Avesta proper.

The deutero-canonical literature is written in ‘Middle Persian,’ or Pahlavi, which was the language of Persia between about the third century B.C. and the ninth century A.D. These Pahlavi writings are an attempt to explain the Avesta which, together with the Zend, is attributed to Zoroaster himself. By this time the Prophet had become regarded as a supernatural being, and his teaching was neglected while his person was exalted. The Dinkart, composed about the ninth century A.D., is a summary and exposition of the Avestan texts and Zend known to the author, who by his quotations exhibits knowledge of a greater number of Avestan Books than we now possess. It is a huge volume of over a thousand pages, although the first two treatises are lost: among the remaining contents is a very fanciful life of the Prophet. The Bundahish and the Zartusht Namah are two later Books, written in Persian, subsequent to the invasion of Islam. The latter bears the date of A.D. 1277, yet purports to give accurate information of the details of the life of Zoroaster. The Bundahish, probably of somewhat earlier date than the Zartusht Namah, is a kind of ‘Enquire within upon Everything’ concerning medieval Zoroastrianism. Among the host of less important Pahlavi Books, only two need here be mentioned—the Arda-Viraf and the Dadistan-i-Dinik. The former is an interesting sixth century account of the nature of Heaven and Hell, and the latter a series of questions put to a ninth-century priest, together with his answers. These deuterocanonical Books form what is known as the Kordah, or little, Avesta. Modern Persian translations of this Kordah Avesta have been published, and are of some use in determining the meaning of words that are obscure in the Pahlavi, but often they paraphrase rather than translate literally.

There are no manuscripts of the Avesta earlier than the thirteenth century A.D., the majority belong to the seventeenth. Not only did the old manuscripts suffer mutilation from Alexander the Great, for the Muhammadans also destroyed many Zoroastrian Books. The Quran was offered as a substitute for the Avesta, at the point of the sword. Although many faithful Zoroastrians refused to accept it, others acquiesced on the ground that many of the doctrines of the Avesta were also to be found in the Quran. A certain number of Persian religious ceremonies were allowed by the Moslem conquerors, and some of the old Fire Temples remained. But the glory of Persia had departed.

Notes

1. Entitled Historia Religionis veterum Persarum eorumque Magorum. Published a.d. 1700.

2. It is not certain which of the three kings who bore this name was responsible for the work.

3. Essays on the Religion of the Parsis, pp. 69-70.

4. Early Religious Poetry of Persia, p. 2.

5. The Living God, pp. 187-8.