Berliner Pahlavi-Dokumente: Zeugnisse spatsassanidischer Brief- und Rechtskultur aus fruhislamischer Zeit.
The book under review represents the first edition of a cohesive group of Middle Persian documents on leather and linen, written in the cursive variety of Pahlavi, emanating from Early Islamic Iran, but presumably reflecting administrative traditions of the Late Sasanian period. They had apparently been found in the vicinity of Qom in uncontrolled excavations, emerged on the antiquities market, and the majority were acquired by the Free University of Berlin. Smaller collections of related texts, published in the same volume, remain in private possession in Los Angeles, Tehran, and other places. A larger group of texts, which likely comes from the same archive, was acquired by the Bancroft Library of the University of California in Berkeley, and scholars are looking forward to its publication by Philippe Gignoux.
As a more remote parallel to this recently emerged archive one can mention the Pahlavi papyri, which stem from the time of the Sasanian occupation of Egypt in the early seventh century A.D., and Pahlavi ostraca from roughly the same period, some of which come from Iran proper. These documents, too, are written in the developed Pahlavi cursive and have predominantly administrative character, but the ostraca tend to be very short, while the papyri are mostly fragmentary. In addition, parts of the Vienna collection of Pahlavi papyri, probably the largest and the most significant one, went to Russia in 1945 as questionable war trophies and remain generally unavailable for scholarly study. Therefore the discovery of longer, well-preserved, and easily accessible texts has the potential for dramatically improving our understanding of Late Sasanian bureaucratic practices.
The core of the volume (pp. 1-201) is the edition, with commentary, of forty Berlin documents, nine documents from Tehran, and six documents from Los Angeles (the first edition of the last group of texts had already been undertaken by Philippe Gignoux). All the documents are published as photographic plates and provided with transliteration and transcription. The remaining part of the volume includes general palaeographic and linguistic notes (pp. 203-28), synopses of datings (pp. 229-34) and personal names (235-36), a survey of the material aspects of the documents, including the technique of their folding, by Myriam Krutsch (pp. 237-48), and a discussion of legal terminology by Maria Macuch (pp. 248-66). The volume has already been acclaimed as a groundbreaking work conforming to the highest standards of philological research. It is my pleasant duty to signal that it has been awarded the Ghirshman Prize of the Academie des Inscriptions et Belles-Lettres and received an honorary mention in the competition for the Ehsan Yarshater Book Award.
From the typological point of view, the edited documents can be subdivided into three main categories (cf. the classification on pp. xv--xiv). The first group of texts contains official orders for, or accounts of, the distribution of payment. Many of the documents belonging to group one feature the formula az an i o bun "--concerning that of the capital," the precise sense of which remains problematic (cf. pp. 257-60). The second group contains receipts issued by private individuals, which are internally referred to as <pOtgl'134>, pgdiray. The apparent external reference to the documents of group two in texts of group one is <ck'>, which Macuch compares with Classical Persian eak 'document, contract', ultimately the likely source of English cheque (pp. 38-39,260-61). The existence of receipts does not imply that private beneficiaries of transfer were literate, since these documents could be countersigned with the personal seals of witnesses. The third, most heterogeneous, group consists of letters, which are frequently recognizable though elaborate greeting formulae.
The documents are dated to the interval of the years 30 and 49, presumably of the Yazdegerd era starting in 632 A.D., which places them in the second half of the seventh century A.D., in the period after the Arab conquest of Iran. Palaeographic data do not contradict this dating, while radiocarbon analysis broadly confirms it (p. xv). It is therefore remarkable that the published texts appear to make no reference to the Arab presence in Iran (cf. the Bactrian documents of the eighth century A.D., which frequently mention the Arabs). One remains under the impression that the old Sasanian administration continued for a while to function essentially independently from the new structures installed by the occupiers. Particularly intriguing is the mention of xwaday i Era n "lord of Iran" in Document 3, whom one is apparently requested to support with a large number of men (p. 15). Is this is a reference to new Muslim authorities or to the old Sasanian dynasty that still claimed loyalty of some of its former subjects? One can hope that the publication of Berkeley documents will shed more light on this historical problem.
Unfortunately, many readings of the texts under discussion remain uncertain for objective palaeographic reasons. The Pahlavi family of consonantal alphabets (abjads) is generally known for the merger of graphemes with respect to their Aramaic prototype, ambiguity that much complicates our understanding of the relevant texts. Inscriptional Pahlavi, typical of Sasanian rock inscriptions and coin legends, has nineteen distinct characters in contrast to twenty-two in Official Aramaic. Christian or Psalter Pahlavi, with its set of eighteen characters, has the additional complication of continuous writing. More radical ambiguity characterizes Book Pahlavi, the form of script used for religious and literary purposes in the Zoroastrian communities of medieval Iran and India. Here the number of distinct signs is limited to thirteen with further mergers in specific positions within a word. Many Middle Persian words recorded in Book Pahlavi require contextual disambiguation, and sometimes their correct readings were not known even to their copyists (thus Indian Zoroastrians used to read the name of Ohrmazd as Anhuma in Middle Persian)
But those who think that the script cannot get any more ambiguous than Book Pahlavi are mistaken: there is also the developed Pahlavi cursive. Its difficulties stem not from further reduction of the number of characters, but rather from their random merger in certain lexemes and forms. To give just one characteristic example, a new reading proposed in the volume under review is <'nwg'byy't> aneiiyad 'of eternal memory' for the epithet that had previously been read as <'nwgbht> anabart 'of eternal destiny' (p. 3). The difference between these two readings cannot, however, be seen in any of the documents published in this volume, where the second aleph is completely absorbed by the following <b>, while the sequence cyy'> is reduced to a dot or a horizontal line. The new reading is backed instead by two Vienna papyri (at least one of which remains unpublished) where the characters of the word are written more distinctly, more or less as they would have been in Book Pahlavi. In practice, this means that no scholar of Cursive Pahlavi can hope to adequately read a particular archive based on studying its palaeography alone. It is necessary for this purpose to have access to the whole corpus of data, including unpublished materials from other archives.
This explains why the pool of researchers who actively contribute to the publication of Pahlavi cursive documents is now limited to two individuals: Dieter Weber and Philippe Gignoux. This should not, of course, be taken as a reproach against these two scholars, who indeed must be admired for devoting considerable parts of their lives to studying texts that may have seemed obscure or boring to other Iranologists. Weber also deserves special praise for the compilation of a sign list reflecting the palaeography of the volume under review, which is meant to contribute maximal transparency to the results of his decipherment (pp. 212-28). But the tradition of interpreting the Pahlavi cursive is obviously in danger, unless Weber and Gignoux succeed in training a new generation of pundits that will take it over.
But how much easier would it have been for the ancient scribes to read the same texts? Of course, they could have the advantage of being Persian native speakers, but the large number of Aramaic heterograms and (pseudo-)historical spellings in Pahlavi must have attenuated it. Correct understanding of the texts apparently depended upon many years of professional training, which would justify the existence of scribes as one of the four hereditary classes in late Sasanian society. Under such circumstances the ambiguity of the cursive script might have even been semiconsciously or subconsciously cultivated as a way of preventing laymen from acquiring scribal skills. Of course, writing systems that become tools of social exclusivity are ultimately doomed, in particular after losing state sponsorship. One is tempted to draw a parallel with the ever-increasing complexity of the Egyptian hieroglyphic system in the Roman period, which went hand-in-hand with the shrinking of its epigraphic community. Thus there is no wonder that the cursive Pahlavi script did not survive the Islamization of Iran and was not retained even among the remaining Zoroastrians, who continued to avail themselves of Book Pahlavi.
In conclusion, I have to mention two cases where I find it difficult to follow Weber's interpretation of the Pahlavi cursive adopted in the volume under review. According to Weber, a striking archaism of this writing system is the grapheme <t>, which is otherwise vestigially present in Inscriptional Pahlavi, but completely absent in Psalter and Book Pahlavi. This hypothesis runs afoul of the general trend according to which the script of the Pahlavi Psalter is always more archaic than the cursive of administrative texts. On the other hand, there is no obvious graphic similarity between the alleged cursive form of <p. (graphically Ti = e.g., <dr>) and its monumental prototype; nor does its distribution have anything to do with that of the Aramaic teth. It is therefore more economical to assume that the alleged instances of <t> in the cursive script represent the simplified form of r <t>, as was traditionally thought. In fact, Weber himself shows that this grapheme can undergo further simplification in the same script, becoming indistinguishable from I <r/w/n> (cf. the spelling of <pt-g> on p. 223). We have already seen that other graphemes can be reduced to a dot, a horizontal line, or simply disappear in individual cursive forms, so the progressive simplification of <t> need not amaze us.
As for the pseudo-heterogram <PWN>, which was used to render Middle Persian pad 'at', its new reading as "corrupt" <N'> in administrative texts would dissociate it in an unwarranted fashion from its antecedent <PWI\I> in Inscriptional Pahlavi. I personally deem it very likely that <PWNI>of the Sasanian monumental inscriptions represents a corruption of the earlier <pt.>, as suggested by Schaeder and others, and that early cursive spellings must have played a part in this reanalysis. I see, however, no justification for assuming that the non-trivial graphic development that must have occurred before the third century A.D. was undone and then nearly repeated itself four centuries later.
ILYA YAKUBOVICH MOSCOW STATE UNIVERSITY / OXFORD UNIVERSITY
|Printer friendly Cite/link Email Feedback|
|Publication:||The Journal of the American Oriental Society|
|Article Type:||Book review|
|Date:||Jan 1, 2012|
|Previous Article:||William J. Gedney's Concise Saek-English English-Saek Lexicon.|
|Next Article:||Sasanian Iran (224-651 CE): Portrait of a Late Antique Empire.|