The Pahlavi version of the Avesta
Middle Persian includes the iranian dialects as they appear from about 300 B.C. to about 900 A.D. They are in general called Pahlavi, which is only the regular development of a derivative of the Old Persian word Parqava 'Parthian'. It is clearer to discuss the dialects partly by dialects and partly by the extant remains.
Arsacid Pahlavi was the official language of the Arsacid dynasty of Parthia, which ruled from 250 B.C. to 226 A.D.; it did not die out with the dynasty. It is represented in some bilingual inscriptions alongside the Sasanian Pahlavi, where it is often called Chaldeo-Pahlavi or Parthian; by the parchment manuscripts of Auroman; and by certain Manichaean texts from Turfan. It is also called Northwest Pahlavi, and apparently was developed from a dialect which was almost or quite identical with that of Media.
The Sasanian or Southwest Pahlavi was the official language of the Sasanian dynasty, which ruled from 226 A.D. until the Mohammedan conquest in 652. It is known from some rock-inscriptions of the kings in the general region of Persepolis, datable in the 3rd and 4th centuries, some being accompanied by a translation into Arsacid Pahlavi or even by a second translation into Greek; from some texts on Egyptian papyri, of about the 8th century; from many religious text preserved by the Zoroastrians; and from some
of the Manichaean texts found at Turfan. In inscriptional form it can be observed in legends of coins, seals and geams, until near the end of the 7th century. It appears to have developed from Old Persian or form a very similar dialect.
The 'Book-Pahlavi' includes the writings preserved by the Zoroastrians of Persia and India, forming a very considerable body of literature divisible into (1) translations of parts of the Avesta, with commentary, (2) texts on other religious subjects, (3) texts on other than religious topics. They represent both Sasanian and Arsacid Pahlavi. They are written in an alphabet derived from that of Aramaic, and, like all the early Pahlavi writings and inscriptions, contain an extremely high percentage of Semitic
words; but many of these were to be read with the Iranian equivalents, even as we write id est and say 'that is', viz. and say 'namely'.
The manuscripts found at Turfan, in the early years of the 20th century, give us texts that are mostly of the 8th and 9th centuries, though some of them go back almost to the beginning of the Christian era. These text represent several dialects, including the Arsacid and the Sasanian types, the Sogdian (known also from a trilingual inscription of Kara-Balgassun), and a dialect known as 'Eastern Iranian', perhaps a derivative of northeastern Scythian, in which there are texts of the Buddhists of Khotan. The
notable peculiarity of these Turfan texts is that they are written in relatively pure Iranian, without the Semitic writings for the words which are to be spoken by the Iranian equivalent.
(Abstracted from : Old Persian, Roland G. Kent, New Haven, 1953)
1) The earliest traces of Pahlavi
To the period of the Sassanian editing of the text belongs the Pahlavi translation and interpretation of the Avesta. At the date when the texts were compiled and edited, the general knowledge of the Avesta and the understanding of the sacred texts was far from perfect. The preparation of a translation or version became necessary. According, the great body of the texts was rendered into Pahlavi, the language used in Persia at the time of the Arsacide and Sassanid. The Pahlavi version and interpretation of the entire
Yasna, Vispered, and Vendidad, with some portions of the other texts, has been preserved. We have not as yet a through enough understanding of this version, as the Pahlavi question is still a vexed one; but as our knowledge of this translation increases, we see more and more its importance. Owing to a somewhat imperfect knowledge of the Avesta texts at the time when the version was made, and owing to the unskilful and peculiar manner in which the Pahlavi translation is made, this version abounds in numerous errors
and inaccuracies. Its renderings, however, are often of the greatest value in interpreting allusions, particularly also in giving hints for the meaning of obscure words, and in such matters it is many times our best and only guide. When more fully understood and properly used in connection with the 'comparative method', reffering to the Sanskrit in interpreting the sacred texts, the 'traditional method' or native explanation is destined to win great results. The 'traditional' and the 'comparative' methods must go
hand in hand.
(Abstracted from : Avesta Grammar, A.V. Williams Jackson , Stuttgart, 1892)
2) The Sasanian inscriptions
As an essential characteristic of Pahlavi writing is the mingling of Iranian with Semitic words, we may perhaps look for the earliest traces of Pahlavi Literature to the legends on the Abd-Zohar and sub-Parthian coins of the third and forth centuries before Christ, quoted by Haug in his Essay on Pahlavi, from Levy's Beitraege in ZDMG. The Iranian admixture is small, but it exists, and its existence is sufficient to show that a species of Pahlavi was already in use and written in Aramaic characters.
Some unexpected information as to the pronunciation of the Iranian element of Pahlavi, in the first and second centuries after Christ, has also been obtained from more complete decipherments of certain legends in Greek characters on the coins of Indo-Scythic rulers of the Turushka dynasty in northwestern India. It had long been noticed that many of these coins bear re-presentations of divinities of a Zoroastrian type, such as mitro, Athro and others; also that the kings' names, written Kanishka
and Huvishka in Indian inscriptions, are wrriten Kanerxi and Oerxi in Greek characters on their coins. It was noticed that the Greek r in these two names was not exactly the same as in some other words, and Burgess suggested that this modified r was used for sh, in the same way as I was used for r in certain cases noted by von Sallet. But it was left for M.A. Stein to show, in 1887, that this modified letter was really not a r, but was borrowed to represent the sound of sh; and, by substituing this sound
wherever this peculiar letter occurred, he was able to decipher several further titles and Zoroastrian names, such as shaonano shao = shahanshah, shaorero = shahrivar, ashaeixsho = Av. Ashavahishta, ardoxsho = ardvaxsho (Av. Ashi-Vanguhi). There also occur on these coins the following Zoroastrian names and words: mithro, etc. = Mihro, Mao = Mah, Arooathpo = Lohrasp, Oado = Vado, Tharro = Farro (Av. Xvarena, np. farr, farra), Oanino = Vanand (Av. Vanainti), Teiro = Tir, and probably Oshlaghno = Varhran
(Av. Verethraghna) and Athsho = Phlv. ataxsh or atash. Allowing for the deficiencies of the Greek alphabet, which occasioned the use of vowel o for the sounds of h and v (English w), in addition to its own, these Greek transcripts ought to represent something like the Iranian pronunciation of the first century after Christ. And it is remarkable that this pronunciation often approaches much nearer to the modern Persian than to the sound indicated by the written Pahlavi, which latter seems to preserve traces of an
older pronunciation more like that of ancient Persian and the Avesta; this is especially seen in the names Mihro, Shahrevaro, and Vado, when compared with the Phlv. Mihr, Shahrivar, and Vato. The final vowel o, which is very rarely replaced by i or e on these coins, is used after every consonant that occurs; but in MS. Pahlavi it occurs only after b, p, t, ch, k, n, and g.
In confirmation of the early corruption of the ancient Persian Mithra into Mihr, the names of Kappadokian months, taken by Benefey and Stern from ancient MSS., may be quoted: though six of these MSS. give the form Mithri, the other four have Mieran, Moar, Muoi, and Muar, showing that the corruption had commenced. The continuance of obsolete forms in Pahlavi writing, even down to the present time, must undoubtedly be a survival due to the conservative instincts of writers; but whether we should be justified in
supposing that Artaxshir was merely such a survival, in the third century, is rendered doubtful by the occurence of his successors' names in the later forms Shahpur and Varhran (Bahram) which show that those kings had no particular prejudice in favour of antiquated language.
3) The oldest Pahlavi writings now extant
For the oldest surviving specimens of actual literary composition in Pahlavi we have to look to the rock-inscriptions of the early Sasanian kings of the third and fourth centuries after Christ. One or two of the shorter inscriptions were copied by Flower in 1667; then by Chardin and Niebuhr; more of them by Ker Porter, Ouseley, and Westergaard; and the whole of them have been copied by Flandin and Coste, and photographed by Stolze and Andreas. De Sacy deciphered the short trilingual inscriptions of Artaxshir-Papakan
(226-241) and his successor Shahpuhar (241-272), on the bas-reliefs of Naksh-i Rustam and Naksh-i Rajab, which consists chiefly of names and titles in Greek and two dialects of Pahlavi engraved in different characters, usually called Chaldeo-Pahlavi and Sasanian Pahlavi. The information thus obtained was soon applied to the decipherment of similar short inscriptions, in Sasanian Pahlavi only, at Shahpur and the Tak-i Bostan, containing the names and titles of Narsih (293-302), Shahpur II (309-379), and Shahpur III
(383-388). It was also applied to the decipherment of legends on the coins of other Sasanian kings. But no further progress was made with the decipherment of the longer Sasanian inscriptions, which contain more information than mere names and titles, until Haug began to apply to the subject the knowledge of manuscript Pahlavi, which we had acquired among the Parsis in India. The best texts, then available, of all the known Pahlavi rock-inscriptions had just been collected by Thomas, mostly from the recent copies made
by Flandin and Coste; and the bilingual inscription of Shahpuhar I, in a cave at Haji-Abad near Persepolis, was selected by Haug as the most complete and legible of the longer inscription. This inscription, which had been very carefully copied and published by Westergaard, is in Chaldeo-Pahlavi and Sasanian Pahlavi, and the two versions mutually assist their decipherment. It begins with the names and titles of king Shahpur, his father, and grand-father, as in his short inscription at the Naksh-i Rajab; and of the
remaining two-thirds of the inscription about one-third of the words in the Chaldeo-Pahlavi version and two thirds of those in the Sasanian version are to be found in the Pahlavi manuscripts of the Parsis. When these words were identified it appeared that the inscription was a record of the kings archery practice in the presence of his nobles, and of his command that, as a spirit's target had been constructed there, the spirit's hand had written that no one else should presume to shoot there after the spirit's arrow
had been shot. As one of the king's title was Minu-chitr min yazdan, of spiritual origin from the sacred beings, it is very probable that the spirit (Minu), mentioned in the inscription, was the king himself.
The remaining inscriptions have not been so fully deciphered, partly owing to their dilapidated condition, or to imperfections in the copies or photographs, and partly owing to difficulties of interpretation. The most complete of these inscriptions is one of 31 lines of Sasanian Pahlavi, near one of the bas-relief of Naksh-i Rajab representing Ahuramazda in the act of presenting a chaplet to the king. The first line, which is difficult to read, seems to introduce the reader to the crown (Kartir) of the divine race,
which, in 1. 27-31, addresses him in the following words:-
[Translation]. That crown am I who made Shahpur king of kings, (his) title of Mobad and title of Herbad have I produced who (was) the crown of Ahuramazda, king of kings, and of Varhran, king of kings, the sons of Shahpur; I who have produced (their) title of Mobad of Ahuramazda (am) the crown of that Varhran, (king) of kings, who is son of Varhran, (even) I who have produced that title of Mobad of Ahuramazda for the saved soul of Varhran.
If this be the correct interpretation of this involved statement, the inscription may probably date from the reign of king Bahram II (276-293). The earlier part of it is chiefly religious in character, as the sacred beings (Yazdan), heaven and hell, the soul religion, sin, and good works are all mentioned. With the aid of extreme patience and a few more photographs, taken when the sunlight is in various positions, the whole of inscription could probably be deciphered, although it contains several engraver's errors.
Two other inscriptions in Sasanian Pahlavi, on the southern portal of the palace of Darius at Persepolis, are also very nearly complete, but not very legible. They were copied by Ousely, photographed by Stolze, and described by Noeldeke in his remarks on the inscriptions in Stolze's Persepolis. The upper inscription cosists of twelve lines beginning as follows:-
[Translation]. In the month of Spendarmat (Esfand) in the second year of the Mazda-worshipping divinity Shahpur, King of kings of Iran and Aniran, of spiritual origin from the sacred beings; on that occasion when shahpur, king of kings, has just now arrived...son of the Mazda-worshipping divinity Ahuramazda, king of the kings of Iran and Aniran, of spiritual origin from the sacred beings...
As this Shahpur is called the son of king Ahuramazda, he must have been Shahpur II, and the date of this inscription is before equivalent to the month of August 311, when he was an infant in his second year. The rest of the inscription will be difficult to decipher from Stolze's photograph. The lower inscription consists of eleven lines, and is dated the month Tir in the eighteenth year of the same king, on the day Ahuramazda, corresponding to 30 November 326 (or February 327). It is rather more legible than the upper
inscription, and appears to refer to the doings and good works of the king, whose name occurs six or seven times.
The remaining two inscriptions are in a much more dilapidated condition. One of them has been by far the longest of the Sasanian inscriptions, originally consisting of about 77 lines, engraved in Sasanian Pahlavi on the rock behind the king's horse in the basrelief of Naksh-i Rustam; but only fragments of 65 lines were sufficiently legible to be carefully copied by Westergaard in 1843. From this copy, and that made by Flandin, an attempt was made by west to decipher one-third of the inscription. The first 34 lines
evidently contained an account of the succession of the early Sasanian kings, from Artaxshir to Bahram II or III, in much greater detail than in the Naksh-i Rajab inscription (5), but with the same particulars of the crown and the title of Mobad and Herbad. So there is every reason to suppose that these two inscriptions are very nearly contemporary.
The other dilapidated inscription is bilingual, consisting of short fragments engraved on detached stones which had fallen from the walls of a ruined building called Paikuli, near Sulimanieh. Each stone supplies short fragments of four to seven successive lines of Sasanian or Chaldeo-Pahlavi, and Rawlinson copied twenty-two such groups of fragments of the former and ten of the latter in 1844; but, as none of them can be read continuously, it is doubtful if more than half the inscription has been found. Transcripts of
these fragments were first published by Thomas in 1868; but, owing to their want of continuity, very little progress has been made in their decipherment. As the names of king Artaxshir and king Shahpur occur, as well as that that of Ahuramazda without the usual royal title, we may conclude that the date of the inscription may have been late in the reign of Shahpur I, or about 270 after Christ. Several names of countries or tribes have been noticed in these fragments, and the kings of some of them are mentioned; also
priests, Ahriman and demons; from which facts it may be supposed that the inscription was both historical and religious.
Besides being contemporary records of the relationship and succession of ten of the early Sasanian kings, these inscriptions are useful guides to the correct reading of many Pahlavi words, where the traditional pronunciation has been led astray by the ambiguity of several letters in the modern Pahlavi alphabet. The Sasanian alphabet being less ambiguous in pronunciation, the orthography of any well-identified word in the inscriptions is a better authority for the reading of its counterpart in the manuscripts than the
so-called tradition of the Parsis, especially in the case of Semitic words.
The Chaldeo-Pahlavi appears to have fallen into disuse towards the end of the third century after Christ; but the Sasanian Pahlavi can be traced, by means of coin-legends and inscriptions on seals and gems, gradually changing its alphabet into modern Pahlavi between the middle of the fourth and the end of the sixth century, though the Sasanian x, sh and final a seem to have been used till the latter end of the seventh century.
4) The origin of the Pahlavi literature preserved by the Parsis
Among the numerous and miscellaneous writings on papyrus, discovered in the Fayum district in Egypt during the last few years, are many fragments of Pahlavi manuscript written probably in the eighth century, being the oldest specimens of Pahlavi writing known to exist. This writing is intended to be similar to that used in the Pahlavi MSS. of the present day, and many words can be read with ease after a little practice; but it is often careless and hardly legible. Some of the fragments appear to contain portions of
daily memoranda of sales or expediture; but, as there are many varieties of hand-writing, the papyri are evidently the remains of a collection of miscellaneous documents.
As specimens of old Pahlavi writing may also be mentioned ten signatures of witnesses on a copper-plate grant to the Syrian Church in southern India, supposed to have been engraved in the ninth century; and four Pahlavi inscriptions, dated A.Y. 378 and 390 (1009 and 1021), containing the names of Parsi visitors to the Kanheri Buddhist caves in Salsette, near Bombay. Although both these specimens are engraved, they must first have been written upon the copper plate and rock, before the engraver began his work. These
centuries after these Parsis left their names in one of the Kanheri caves, the oldest surviving Pahlavi MS., preserved by the Parsis in India, was written at K'ambay in 1323.
From the foregoing sketch of the progress of Pahlavi writing it will be seen that the modern alphabet is a direct descendant of the Sasanian alphabet, the changes of form in most of the letters having been made chiefly between the years 350 and 680, though the Semitic final a, which is written like mn in modern Pahlavi, is often like m on the Fayum papyri. With regard to changes in the written language, hardly any of the Semitic words used in Sasanian Pahlavi have disappeared from the modern MSS.
(manuscripts); Iranian suffixes to Semitic verbs and particles began to appear in Sasanian Pahlavi about the year 270, but the crude form of the Semitic verb continued in use for the past participle and preterit, the suffix -t or -d converted it into the third person present, singular or plural, and the suffix -e, either alone or in addition to -t or -d, appears to give a conditional meaning to the verb; while the suffix -m of the first person singular was used only with the
Semitic verb hava and some particles in the inscriptions.
There is every reason to believe that an extensive Pahlavi literature, in all branches of knowledge, had come into existence before the end of the sixth century; and, although the troublous times which then followed, and soon led to the Arab coquest of Persia, must have checked all literary pursuits, we shall find that Pahlavi works were still being written as late as the end of the ninth century. In the mean time the modern Persian alphabet had been invented, and the use of Pahlavi was then soon confined to the Parsi
priests, who seem to have continued to make additions to some Pahlavi works, such as the Bundahish, till the end of the eleventh century. Since that time, Pahlavi writers have rarely attempted to compose anything in Pahlavi beyond invocational introductions and colophons to the manuscripts (MSS.) they have copied, with probably some attempts at translating short Avesta texts. They have also prepared Pazand or Parsi versions of old Pahlavi texts, in which the Pahlavi words are transcribed in Avesta or modern Persian
characters; the Semitic portion of them being always replaced by their Iranian equivalents. As the priests diminished in number, they found it impossible to provide sufficient copies to keep all their religious MSS. in existence; far less could they attempt to reproduce MSS. on other subjects. Under these circumstances much of the old Pahlavi literature was, no doubt, lost between the ninth and twelfth centuries.
The Parsi priests who emigrated to India in the eighth century had probably very few MSS., and most of these they, no doubt, lost before the end of the twelfth century, as we find, from colophons preserved in various MSS., that they received from iran a copy of the Vendidad with Pahlavi in 1205, and one of the Arda-Viraf-namak after 1269, about which time several other MSS. were probably brought from Iran by an emigrant priest. Copies of such imported MSS. were certainly made in India between 1320 and 1325, as will be
seen hereafter (18, 19, 58, 97, 101). But it appears, from letters received from Iranian priests in 1478, that the Indian Parsis had then become ignorant of Pahlavi and wanted information on many subjects. Copies of these letters from Iran, and of other received on fourteen later occasions, down to 1673, are preserved in the Persian Rivayats of the Indian Parsis; and it appears from some of them that a Paz. Yosht-i Fryan was sent to India from Iran in 1553, and MSS. of the Visperad and Vishtasp Yasht from Kirman in 1627.
During the eighteenth century the Nirangistan and Denkart first reached India, and they have been followed by the Datistan-i Denig, the Iranian Bundahish, and other important texts during the present century; so that the Parsis in India now possess copies of all the Pahlavi literature known to exist, and their priests are beginning to understand it far better than any of the Iranian priesthood.
This Pahlavi literature may be divided into three classes. First, Pahlavi translation of Avesta texts, intermingled with Pahlavi commentary. Second, purely Pahlavi texts on religious subjects, or connected with religion. Third, Pahlavi texts on miscellaneous subjects not intimately connected with religion. And it will be desirable to describe the texts in each class separately, so far as information has been obtained, beginning with the longest, and dwelling chiefly on those texts which are least accessible.
Abstracted from : Pahlavi literature, E.W. West,
in Grundriss der iranischen Philologie, II. Band, Wilh. Geiger und Ernst Kuhn, Strassburg, 1896-1904
Page: 1 2 3 4 5 6 Next Top Home
Copyright © 2001-2014, Farvardyn Project
Optimized 1024X768 with Internet Explorer 5+