David Neil MacKenzie, the dedicatee of this volume, died in 2001 after a lifetime in Iranian studies, first as Lecturer, later as Reader at the School of Oriental and African Studies at the University of London, and finally as Professor at Göttingen, where he spent the last 20 years of a distinguished academic career. A strong personality who seemed to reck little of friends or enemies – and he had many of both, including the present writer – he tended to rouse strong feelings amongst his acquaintances; his students learned what to expect from him, and colleagues were also under no illusions. It is fitting that a large group of them has been gathered together by Dieter Weber for the 25 essays in this volume, forming a very imposing and interesting tribute to an important scholar whose many contributions to Iranian studies spanned a half-century. (These are usefully itemised in a 13-page bibliography).

The volume comprises a quite exceptional collection of essays, all of them interesting and significant, and many of them outstanding. I should have liked to discuss each of them individually, but that would greatly exceed the space allotted to this review. I shall therefore have to confine my remarks to short notes on particular articles without in any way implying judgment of others, which I shall only mention.

In the first essay Carlo Cereti, starting from J. P. de Menasce's 1945 edition of the polemical Pahlavi Škand Gumānīg Wizār (‘Doubt-destroying Exhibition’), presents a new discussion of its presumed author Mardānfarrox's demonstrated familiarity with Manichaeism, and includes a new translation and commentary on its difficult sixteenth chapter.

The next essay, by Iris Colditz, discusses some surprising Zoroastrian terminology used in the Manichaean Middle Persian Šābuhragān.

Everybody knows the special place of the date palm all over the Middle East, including non Islamic communities too. Almuth Degener, once a student of MacKenzie's, notices a striking echo of the Pahlavi poem Draxt ī Asūrīg (‘The Assyrian Tree’) in a popular account published in Tehran in 1361/1942 ‘The Date Palm’ by H. Morādi-Kermani.

It appears that in 1958, in a letter to Stig Wikande, the historian of religious Mircea Eliade wrote that he despaired of anyone who thought that the study of onomastics had anything to contribute to religious history. Such nonsense would not be worth mentioning were it not cited by Gignoux in his brief but interesting note on some of the theophoric names of the Sasanian period in Iran. Amongst names which refer to belief, anōš ‘immortal’ was quite common, whereas farr ‘glory, prosperity, etc.’ went everywhere (cf. Holophernes). Gignoux includes statistics on the number of occurrences of the names.

In a densely annotated note on Avestan geography which attempts to clear up some of the unclear points of the well-known “Khwarezmian Hypothesis” for the location of the Avestan Airyana Vaēďjah, Gherardo Gnoli does not in fact advance matters very much beyond what he has previously written. Everyone agrees that the Airyana Vaē ah, the ‘Wide Expanse of the Aryans’ was the original home of the Avestan speakers because the Avestan texts say so. Whether it was also Zoroaster's original home is probable, but not proven. The problem is to locate A. V. somewhere, and it was Markwart who first (1901) suggested on geographical grounds, supported by some doubtful history, that it might be ancient Khwarezm. This notion was received with approval by some very serious Iranianists, amongst them Bailey, Benveniste and later Henning (in his Ratanabai Katrak lectures of 1949). Of course the problem cannot be separated from the very vexed question of the date of Zoroaster, about which there is still no real consensus amongst experts, and the whole matter has rapidly spun out of control. Those who plumped for historical Khwarezm had rapidly to amend their ideas to an unhistorical “Greater Khwarezm” as Henning did, and others such as Gnoli, who was attracted to a “Sistan Hypothesis”, had to amend it in turn to a sort of “Greater Sistan” which extended north east across present-day Afghanistan so that, as MacKenzie remarked later, both areas partly overlap. Henning added strength to the Khwarezmian Hypothesis by adducing material from his study of the Khwarezmian language (in which he was the first expert) which he thought exhibited a special relationship between it and Avestan. Gnoli, although he mentions MacKenzie's contributions to this debate, especially his 1988 lecture “Khwarezmian and Avestan”, in which MacKenzie examined in detail all of Henning's linguistic examples of a supposed especially close Khwarezmian-Avestan relationship, does not mention clearly how MacKenzie effectively demolished them all, except one. This one example, Gathic Avestan tkaēša- ‘false doctrine’ is paralleled elsewhere in Iranian only in Khw čkyš- ‘a lie’. This word is of course closely related to Av kaēš- ‘teach’ (and Middle Persian kyš ‘false teaching’, but New Persian kīš ‘faith’) showing uniquely in Av t- and Khw č- as some sort of negative prefix, and Khw č-<*ty-,mirrored in the peculiar Av spelling t-. But even this special connexion is lost of in fact both negative prefixes <*duš, common all over Iranian. In view of Khw δmn ‘enemy’ <*duš-man-, MacKenzie at first rejected the notion of a Khw č-<*duš-. But MacKenzie told me that Nicholas Sims-Williams has pointed out that *duš-kaiša- is still a candidate, by comparing e.g. M Sogd ckšty ‘ugly’ <*duš-karštaka-. Henning was always careful to distinguish conclusive proof from proof by linguistic evidence alone, and at present, after the demolishment of Henning's last example, there is no linguistic evidence for the Khwarezmain Hypothesis either. The whole problem is immeasurably complicated if the date of Zoroaster (or indeed the Indo-Iranian homeland problem) is included; no likely agreed answer to any of these questions is in sight, though perhaps at present those who favour a date for Zoroaster of earlier than 1000 bce (e.g. Mary Boyce, Humbach, Gnoli and, for what it is worth, the present writer) against those who favour the ‘traditional date’ of 258 years before Alexander (e.g. Henning, Gershevitch, and many others) are slightly in preponderance. There is, at present, simply not enough evidence of the right kind for a convincing decision.

In a major study, Philip Huyse examines the probable origin of the final stroke commonly appended to words in the Phl script, proposing to connect it to the ‘final –y’ of Middle Persian inscriptions (first studied by Henning in Mitteliranisch, in 1956), to my mind convincingly. The problem has a long history since 1956 in which much ink has been spilled. Building on an article by Walter Belardi written in 1986 in which purely graphical grounds for the Phl final strokes were adduced, Huyse proceeds to examine certain MP inscriptions where the reason for the ‘final –y’ is much less clear, as Henning began to do. Huyse has had the brilliant idea of comparing very late OP inscriptional practice, where as is by now well known very substandard linguistic procedures take over; there was a great decay of the elaborate OP morphology in speech to what was, by say 400 bce, already early Middle Persian; this led to false writings in very late OP inscriptions by people who no longer knew the language (full material in Rüdiger Schmitt 1999, Beiträge zu Alterpersischen Inschriften). Tracing the details of exactly what must have happened makes fascinating reading, and – to my mind convincingly – he is even able to set dates.

It appears that interest in Yezidi studies is increasing in the West, and a new study of its sacred literature is provided by Philip Kreyenbroek, continuing his work on the subject.

In ‘structure d'actances’ in modern Iranian languages, Gilbert Lazard provides a much-needed survey and summary of the present state of research on ‘ergativity’: I use inverted commas because the word is used to mean very different things by different scholars. Lazard distinguishes between the type OP taya manā krtam ‘this is what I did’ and utā-šām ahura-mazdā naiy ayadiya ‘and they did not worship Ahura Mazda’, examining the situation in Modern Persian, Kurdish, Taleši and many other West Iranian languages, as well as, for East Iranian, Munji, Pashto, and some Pamir languages. It is a pity that he does not treat Balōči, which presents by itself a nearly typical picture of the whole family: ‘ergativity’ is strongly characteristic of some dialects, others have gone over entirely to the ‘active’ construction of New Persian, whilst still others show examples of every kind of ‘mixed’ construction. Balōči dialects are now well-described in the literature.

Maria Macuch writes a study of Sasanian legal documents, with many examples, both grammatical and syntactical. The study is to be complemented by her forthcoming book on Middle Persian Legal Terminology.

In 1970 MacKenzie published a re-edition of the Sutra of Causes and Effects in Sogdain, a basic Sogdian text previously edited by Gauthiot. As B. and D. Meisterernst remark in their extensive new study, this Sogdian text, together with the Vessantara Jataka, still form the basis of study of B Sogdian. MacKenzie's edition itself marked a great advance on Gauthiot's, mainly because of MacKenzie's greater use of the Chinese original. Clearly, however, the last word had not been said about this important text, and in their new study the Meisterernsts examine in still greater detail the textual structure of both the Chinese and the Sogdian versions, making much fuller use of the Chinese than was possible for MacKenzie.

In a study of the Sasaniah Pahlavi names for the chess pieces, Antonio Panaino provides a convincing unscrambling of a very corrupt passage in the Pahlavi text Wizārišn ī čatrang (‘Explanation of Chess’), namely in Chapter 10. A main difficulty is to resolve the reading of what looks like m'tgdnl'n, perhaps ‘ministers’ and equivalent to English “rook”, which incidentally is given in English dictionaries, unconvincingly, as deriving from an obscure Arabic ruqq. Could it not rather be connected with Prs rux “cheek” (> “side”), as its seems MacKenzie has suggested, and exhaustively discussed here?

Ludwig Paul's study of the dialectical origins of the vocabulary of the Šāhnāme_partly continues the work of Lenz (in ZIII, 1926). He seems to accept A. Lubotsky's suggestion (in fn. 10) that Persian farr ‘glory, etc’ (<Iranian *farnah-) is a loanword from some Eastern Iranian language (Scythian?) rather than the traditional loanword from *Median. But in my opinion it is neither; it is genuine Persian, as I have suggested in an article which attempts to explain Av xvaranah- as well, in the Humbach Festschrift Philologica et Linguistica, 2001, “Splendour and Fortune”, pp. 485–486). Linguistic study of the Šāhnāme has hardly begun.

In ‘Reste einer soghdischen Version von Huyadagmān I’, Christiane Reck provides a new translation and detailed commentary. (Incidentally, it was MacKenzie who provided the correct reading of what Mary Boyce read as Huwīdagmān in her Hymn Cycles in Manichaean Parthian of 1954), and a commentary on Mary Boyce's edition is given. Some good manuscript photos are supplied.

Martin Schwartz's ‘On the Khwarezmian Loss of *-r-’, although very brief, is convincing, adding an important note to Khwarezmian historical phonology.

Shaul Shaked provides a short list of Iranian words newly found in Aramaic, as well as convincing corrections to previous suggestions. To be especially noted is his discussion of Aramiac šqwrt' occurring on a plaque in the Foroughi collection, most likely ‘drinking cup’ or the like, which turns out not to be Iranian, despite previous attempts to connect it with ‘hunt’ words, e.g. Persian šikār.

Dan Shapira, in some difficult Bundahišn passages relating to flowers, discusses some distinctions, with a full commentary on some quite obscure flower names.

In a major study of the Bactrian calendar Nicholas Sims-Williams and François de Blois, both leading scholars, have collaborated to present a study which is the fruit of much research. Bactrian studies have been revolutionised since the discovery in Afghanistan of caches of documents, one of which has become the Khalili Collection. Sims-Williams has been their principal interpreter since the 1990s. New day and month names have been discovered, lists of documents and inscriptions have been made, dating formulae are devised, Beruni's Chronology is re-examined in detail. Many of the names are better explained as well.

In ‘Avestan III’ Skjaervø examines anew the wealth of endings attested in various Avestan mss for the loc. sg. of u-stems. He shows that the –o ending of a few Yasna mss is to be discarded, at best the result of analogy used by scribes. Other proposals are made and closely argued.

Pēsūs is a rarely-occuring female demon noticed by Mary Boyce in one of the Manichaean Parthian fragments from Berlin. She tentatively identified her with the Syriac Nebrō'ēl, the demonic partner of the male Šaqlūn, with whom she bore the pair Adam and Eve, the first human couple. Werner Sundermann shows that Pēsūs/Pēsōs is certainly in Parthian, from whence he came into Sogdian. He corroborates the identification with Nebrō'ēl as well as a connection with the demon Āz ‘envy’. He also mentions Middle Persian afsōsag, very likely a derivative of afsōs in an earlier meaning ‘complaint’.

In ‘Eleven Etymologies’ Finn Thiesen discusses some Persian words. I was particularly struck by his clever explanation of the prefix xar-‘big’, where everyone thinks of ‘ass’, even though it is not at all clear how xarbuz ‘melon’ or xarmast ‘dead drunk’ can contain xar ‘ass’. The only reasonable example to justify such a folk etymology is the well-known xargūš ‘hare’ (‘ass-ear’), paralleled in Balōči by kargōš ‘id’, from which an expected Bal *kar has been extracted. Other Persian examples such as xarsang ‘large uncut stone’, xarmagas ‘horse-fly’ (not ‘ass-fly’) etc. do not seem to show any possible connection with xar ‘ass’. But Thiesen has had the clever idea that xar- ‘big’ is still the same word as xar ‘ass’ etymologically. The connection becomes clear through Persian xarvār ‘ass-load’ (Ir *xara-bāra-), roughly 300kg., which become early generalised to ‘large load’ whence a prefix xar- ‘large’ was abstracted. Persian xarčang ‘crab’ is similarly made from ‘large claw’, here really folk-etymologised from Pahlavi karzang.

It takes a great deal of specialisation to even attempt the readings of the Pahlavi ostraka offered by Dieter Weber, and his study of them fills in a lot of gaps with readings of 31 of them. Photos are provided.

The first edition of the Sogdian ‘Ancient Letters’ was published by Reichelt in 1931; since then much work has been done on them by e.g. Henning, Harmatta and Sims-Williams. In her remarks on them Antje Wendtland offers a new discussion of some of the problems of their interpretation.

That there are fragments of the Mahāyāna Samghāţasūtra also in Sogdian is perhaps not known to everyone, but as Yakubovich and Yoshida point out, there are no less than 6 mss, some in Berlin and some in St Petersburg. Not one is extensive, in no way comparable to the huge Khotanese ms of the sutra. In this thorough and most impressive study, they pose the parameters for deciding the origin of at least some of the Sogdian version(s) from the Sanskrit. The commentary on the text-fragments is extensive, accompanied by a complete glossary and 8 pages of excellent photographs.

In ‘The Tati Dialects of Kalāsur’, Ehsan Yashater provides information about a hitherto undescribed Tāti dialect; but no texts are given, no glossary, no bibliography and no map. The article seems to be a left-over from Yarshater's previous extensive studies of Tāti dialects, cf. e.g. his Grammar of Southern Tāti Dialects of 1969.

There is a fair fund of Arabic and modern Persian words to be found in the Old Uighur material found in Turfan and Dunhuang, here studied by Peter Zieme, and he lists some of them, together with a full commentary.

All articles except that of Yarshater provide full bibliography.