Frahang I Pahlavīk . Edited, with transliteration, transcription and commentary from the posthumous papers of Henrik Samuel Nyberg by Bo Utas, with the collaboration of Christopher Toll. pp. xviii, 175, 2 pl. Wiesbaden, Otto Harrassowitz, 1988. DM 74.

Although a version of the Frahang TPahlawTg has been known to Western scholars since 1771 (rather than 1711, as stated on p. vii), the difficulties of some parts of the text have long proved insurmountable. Iranists have been looking forward, therefore, to this edition of the Frahang. The late H. S. Nyberg, renowned both as an Iranist and a Semitist, was eminently qualified to undertake this important task, which, sadly, he did not live to complete. His work is based on a careful scrutiny of all important sources, including texts which are arranged according to subject and those which follow an orthographical order. After Junker's outline of 1912, a revised textual history was also envisaged, but only a beginning of this was found among Nyberg's papers. Nyberg, it seems, had begun his work on a new edition of the Frahang in the 1930s, but he was repeatedly forced to lay it aside. At his death, he left a number of handwritten versions of the text, the last of which is photographically reproduced in the book, as well as 56 pages of transliteration and transcription. Only a small part of the Commentary was ready, and the Editor had to draw extensively on notes from Nyberg's lectures, on Nyberg's files, and on his Manual ofPahlavi. Revised sets of references to occurrences of the Aramaic words found in the Frahang were supplied by Professor Toll. The posthumous edition of the Frahang bears testimony both to Nyberg's vast learning and to Utas's dedication and editorial skills. The work makes the Frahang much more accessible, and solves many riddles. It will be warmly welcomed by Iranists and Semitists alike. Inevitably, in a difficult text where context can hardly offer any guidance, points of debate remain. Until further evidence of its existence can be adduced, for example, a Phi. *azbay " real" (deriving, it is suggested on p. 73, from the unlikely Olr. *astbava-) remains very doubtful. The conjecture is based on Nyberg's reconstruction of ch. IX (pp. 7-8), where he emends TTMH to * TYRH (for * T'RH = T'LH), and reads * T'RH KZB' " false jackal", which he contrasts with T'LH 'zb', found earlier in the text. It is suggested that 'zb' is the antonym of KZB'. Although the original TTMH (so both Haug and Junker), may well be wrong, no variant *TYRH is mentioned in the critical apparatus, so that it is not clear how Nyberg arrived at his reconstruction. Given the similarity, in Pahlavi script, between KZB' and 'zb', stronger evidence would be required to make Nyberg's explanation seem probable. The interpretation of ch. XXX. 17/ given in the Commentary (p. 110), is simply not convincing: hamak-vanaktom gavak-pil; danb zanb; giya e dazd sikkar appdr, "the (most) allvanquishing = having a father engendering increase; shore = id.; burning (old injunctive 3 sg. of dazitan) the grass (makes) the porcupine (go) away". Several more plausible explanations could be put forward, but none appears to be entirely satisfactory. Attention should be drawn, however, to the fact that the ideogram for MP. duzd, "thief", is GNB\ often written znb'\ this may help to explain the sequence dnb ( = *gnb), znb, gb' 'y dzd. Here and in ch. X.37 (p. 77), the particle 'y, ay "that is to say", appears to have been misunderstood. Instead of Nyberg's reading "*x"ar sikkar, porcupine (?)...(*x"ar = xarl)" on p. 109 (ch. XXX. 12), one might perhaps think of Phi. xwalist, MMP. xw'ryst, "sweetest", and understand ski as sakar, "sugar". What looks like W KR' (p. 37, no. 94: p. 119), glossed in Arabic script y'ny w'a"m, baffled Nyberg, who associated w'd'm with NP. badam, "almond". In view of the possible meaning of