CHAPTER 1
The history of Alexander
Is all an ancient fable now.
– Go bring on what is novel, new,
For novelty's more savoury.
Farrukhī Syīstānyī (d. 1037)1
The Greek Background
It seems that Persian and Arabic authors of the Islamic period had no access to so-called historical sources on Alexander the Great (that is, the apparently historical accounts of Arrian, Plutarch, Diodorus, Curtius and Justin – some of which are not historical, at least in a straightforward sense).2 The Arabic and Persian sources that deal with Alexander's history are based on a Greek work of popular literature known as the Alexander Romance. It is also known as the Pseudo-Callisthenes because the work has erroneously been attributed to Callisthenes, the historian in Alexander's court, on the grounds of several fifteenth-century manuscripts.3
The formation of the Alexander Romance was a gradual process. The traditional view is that the composition of the work as a single entity did not take place until the third century AD, shortly before its translation from Greek into Latin by Julius Valerius.4 The anonymous author of the Romance is believed to have been ‘a competent speaker and writer of Greek’ from Alexandria.5
The textual history of the Greek Alexander Romance is a very complicated one. The Greek work attributed to Pseudo-Callisthenes that appeared in the Middle Ages survives in three major versions known as ‘recensions’:6
(1)The α-recension: represented by a single manuscript7 dated to the eleventh century AD.8 It is the source of the Armenian version9 (c.AD 500)10 and the Latin version by Julius Valerius (c.AD 340).11
(2)The β-recension: the author of the β-recension wrote some time after the Latin translation of Julius Valerius (c.AD 340), but he was apparently unaware of the variants in the Greek source of the Armenian version.12 The β-recension is represented by three ‘sub-recensions’:
a.Sub-recension ε: (MS Bodl. Barocc. 17, thirteenth century AD): with a strong interest in Judaism (it contains the visit to Jerusalem).13
b.Sub-recension λ: a variant of the β-recension, preserved in five manuscripts. The most substantial additions are in Book III.14
c.Sub-recension γ: the longest of the Greek recensions. It follows the basic structure of α and β, but incorporates new material from ε.15
d.Manuscript L (Leidensis Vulcanianus 93): a unique variant of β with more adventures, in particular the episodes of the diving bell and the flying machine (II, 38–41).16
(3)The δ-recension: not represented by any Greek manuscript. However, it is generally believed to be the source of oriental versions (in particular Syriac, Arabic, Persian and Ethiopian). It is also the source of the tenth-century Latin translation of Leo the Archpriest, known as Historia de Proeliis.17
Table 1.1 Variants of motifs in later versions of the Alexander Romance18
The motifs in Table 1.1 are those that are only added to later versions of the Romance. The table shows which recension the later versions are based on. For instance, the twelfth-century Persian poet Niẓāmī is the only author who included the motif of mermaids dancing on the banks of a lake at night.21 This leads us to conclude that his source was probably based on sub-recension ɛ or γ. Another example is the Ā'īna-yi Iskandarī (Alexander's Mirror) of Amīr Khusraw of Delhi (d. 1325), in which the motif of the diving bell appears.22 Thus this implies that Amīr Khusraw had access to an ultimate source similar to manuscript L or sub-recension λ.
Syriac Sources Relevant to this Study
The most widespread Arabic and Persian traditions about Alexander are adaptations of three Syriac works – the Romance, the Legend and the Poem – which were translated at unknown but evidently early dates.23 Another Syriac text, the Laments of Philosophers at Alexander's Funeral appears as an integrated part of the tale in Arabic and Persian versions of the Alexander Romance, although it was not included in the original version. A further Syriac text is the Khuzistān Chronicle, and while it is not related to the Alexander Romance, it demonstrates the Nestorian influence at the end of the Sasanian period, when the Alexander Romance was introduced into Persian literature. It is believed that all these textual traditions became commonplace in the Syriac context within Christian communities from the seventh century AD onward.24
The Syriac Alexander Romance (Taš‘ītā d-’Aleksandrōs)
Without doubt, the Syriac version of the Pseudo-Callisthenes is the most influential of all oriental versions of the Alexander Romance, harking back to the seventh century AD.25 Wallis Budge edited and translated it into English in 1889.26 It consists of three sections that coincide with the Greek textual tradition of the Pseudo-Callisthenes. The source text was related to the Greek α-recension but differs so considerably that it has generally been reckoned to be evidence of a lost Greek recension known as the δ-recension, mentioned above. The first section contains 47 chapters, the second only 14, and the third has 24 chapters.27
There are two important aspects of this version relating to the Arabic and Persian sources. Firstly, the Syriac text offers a fairly complete and accurate account of Olympias's affair with the Egyptian pharaoh Nectanebo, as related in the Greek version, which is not found in most Arabic and Persian sources. Secondly, it describes the expedition carried out by Alexander beyond the River Oxus in Central Asia and China,28 including his visit to the Emperor of China and his adventures there (such as the dragon episode), which are also standard features of the Persian versions.29
The Syriac Alexander Legend
The Exploits of Alexander (Neṣḥānā d-’Aleksandrōs), translated into English by Budge as A Christian Legend Concerning Alexander, is a short appendix attached to Syriac manuscripts of the Alexander Romance.30 It was probably composed by a Mesopotamian Christian in Amid or Edessa,31 and was written down in AD 629–30 after the victory of Emperor Heraclius over the Sasanian king Khusraw II Parvīz.32
In this Christian Legend, Alexander becomes a Christian king who acts through God's will. The most important role of this text in the development of the Alexander Romance is the fact that the fusion of the motif of Alexander's barrier with the biblical tradition of the apocalyptic people, Gog and Magog, appears for the first time in this text.33 The story of Gog and Magog and the Gates of Alexander became a very important component of Arabic and Persian sources.
The Syriac Alexander Poem: A Metrical Discourse (mēmrā) on Alexander Attributed to Jacob of Serūgh (d. 521)
The Christian Legend was the source for a metrical homily entitled ‘Poem on the pious king Alexander and on the gate, which he built against Gog and Magog’.34 The Poem was probably composed in around AD 630–40 by an anonymous Christian author35 in northern Mesopotamia, probably in the neighbourhood of Amid.36
In this text, Alexander appears as a wise and pious king who is only God's instrument in his divine plan. This is said to be the text alluded to in the Qur'ān, which explains how a pagan conqueror managed to be praised in the Muslim holy book.37
The main content of the Alexander Poem deals with Alexander's travels to the Land of Darkness and his search for the Fountain of Life. Here Alexander starts his journey in Egypt and, after sailing for four months, he arrives in India, where the Fountain of Life is. Alexander's cook manages to bathe in the fountain when he goes to wash fish in it, but Alexander himself does not obtain immortality because he does not find the fountain. The other important element of the Poem is the ‘brass and iron door’ that Alexander builds to enclose Gog and Magog. Both motifs are important components of the Arabic and Persian versions of the Alexander Romance.
The Laments of the Philosophers over Alexander in Syriac
In the early Middle Ages, collections of sayings of various ‘wise men’ came to be attached to the story of Alexander's death, and in the course of time these gained enormous popularity both in the East, where they originated, and in the West, translated from Arabic.38
This text, which was originally an independent work, was subsequently incorporated into the Alexander Romance cycle. The Arabic texts of Ya‘qūbī (d. 897),39 Eutychius of Alexandria (d. 940),40 and Mas‘ūdī (d. 956),41 and the Persian Shāhnāma of Firdawsī (completed in 1010)42 contain the Laments. Furthermore, it is likely that the Laments were an integrated component of the Alexander Romance by the tenth century AD.
The Khuzistān Chronicle
In 1889 the Italian scholar Guidi edited an eastern Syrian chronicle that covers the late Sasanian and very early Islamic period. Four years later, Theodor Nöldeke translated the text into German, and dated it to the late seventh century AD, arguing that it was probably composed in around 652 or very soon afterwards, because the last reported event is the death of Yazdgird III, the last Sasanian king (AD 632–51). He also argued that its provenance was southern Iran.43 Thus, the text is known as the Khuzistān Chronicle.44 It fills 20 folios at the end of a manuscript (Borg. Syr. 82) of the Synodicon Orientale, a collection of the records of councils held by the Nestorian Church in Persia.
This single and very brief Syriac text is not related to the Alexander Romance. However, its contribution to knowledge of the present subject is considerable in two respects: firstly, it shows the position of extraordinary influence achieved by Yazdīn, Khusraw Parvīz's (r. 590–628) Christian minister, which indicates the Christians’ power in the Sasanian court in the seventh century AD;45 secondly, it contains some brief information on Alexander that demonstrates what was known about Alexander the Great in Persia in the seventh century AD. The first passage is prefaced by a brief description of Merv46 and a note about the conquests and death of its founder, Alexander the Great. The passage, according to Guidi's Latin translation,47 is as follows:
[Merv] was built by Alexander, son of Philip, and it was called Alexandria. Having defeated and subjugated many peoples of the East, he undertook to return to his country, but he was poisoned by his servants, [near] the Euphrates, at a place called Bāgniqyā in the Babylon area. He reigned twelve years and six months.48
The text also informs us that Alexander built Alexandria in Egypt on the advice of his tutor, Aristotle.49 This kind of brief reference to Alexander's deeds and death can be found in the sources of the Islamic era too. He was particularly recorded as a founder of cities in Central Asia; for instance, the historian Ṭabarī also mentions the names of the cities that Alexander founded all around the conquered territories.50
The fact that Christians had influence in the court at the end of the Sasanian period is important because it shows that the Alexander Romance survived and was introduced into the Persian tradition thanks to the Nestorians. Some scholars51 have doubted the existence of a Middle Persian version of the Pseudo-Callisthenes for ideological reasons. They base their theory on the Pahlavi literature, a faithful mirror of the political and religious ideology of the Sasanian era, which depicts Alexander as the ‘cursed one’ (gujastak in Pahlavi) and equivalent to the greatest enemies of Irānshar (that is, Ḍaḥḥāk and Afrāsīyāb).52 Thus, they argue that the Sasanians would not have tolerated the translation of a text that represents one of the greatest enemies of Iran so positively. However, the presence of Christians and their influence in the Sasanian era53 can shed new light on the development of the Pseudo-Callisthenes tradition in Persia before the Arab conquest. In addition, Byzantine and Arabic sources54 bear witness to the existence of Greek works (in particular scientific and philosophical texts) that were translated into Middle Persian during the late Sasanian era, so it is virtually certain that they existed, though none of these works has come down to us.
Furthermore, the development of the Alexander Romance in the pre-Islamic Persian tradition needs to be studied by taking into account both historical elements besides the content of Arabic and Persian sources, which are based on Sasanian sources.
The Development of the Alexander Romance in the East
Theodor Nöldeke laid the foundations for research on the Eastern versions of the Pseudo-Callisthenes tradition. His most important contribution was his explanation, through a philological examination of the Syriac text, of what led to the development of the literary forms of the Alexander Romance found in Syriac, Arabic, Persian and Ethiopic literature. Nöldeke proposed the theory that a Pahlavi version must have preceded the Syriac version of the Pseudo-Callisthenes because the Syriac text contains some discrepancies with the Greek Pseudo-Callisthenes, and also a certain number of errors in the Syriac transliteration of Greek proper names.55 He dated the lost Middle Persian intermediary to the late Sasanian era (c. sixth–seventh century AD).
Nöldeke's argument, which claims that the Syriac version must have been based on a lost intermediary Middle Persian translation of the original Greek, has been generally accepted ever since. However, Richard Frye56 cast doubt on Nöldeke's theory by arguing that the Syriac version probably descends from a much older Syriac translation of the Greek Pseudo-Callisthenes. He based his theory mainly on three points. Firstly, the philological evidence does not reflect a Pahlavi source but might be the result of references to tales from folklore or oral tradition that were probably current throughout the Middle East and the Iranian world. Secondly, the Syriac version of the Pseudo-Callisthenes makes no mention of Alexander having Persian ancestry – this is mentioned by Ṭabarī and Firdawsī, and legitimates Alexander by making him the stepbrother of Darius III. Lastly, for ideological reasons a Middle Persian version of the Pseudo-Callisthenes could not have existed.
Nöldeke's theory has also been challenged by Claudia Ciancaglini, who believes that the Persian influence in the Syriac Pseudo-Callisthenes must be ascribed to Neo-Persian rather than Middle Persian. She argues that:
Part of the history of the Syriac Pseudo-Callisthenes takes us back to northern Iraq, where we know that, at least before around 1500, the predominant language of culture was Persian. The Oriental Christians in this region addressed a public that was educated in Persian, not in Arabic: it is therefore highly likely that quite a large number of Persian elements, especially the glosses, entered the text long after the time of the first translation from the Greek.57
Kevin van Bladel does not accept Ciancaglini's arguments against Nöldeke on this issue.58 He believes that ‘the only explanation for the confusions in Greek names by the Syriac translator is to propose that there was a Middle Persian antecedent in Pahlavi script’.59 In the same context, it is worth mentioning the argument of Monferrer-Sala, who also believes, for textual reasons, that ‘the Syriac version cannot be considered a translation of any of the Greek PC [Pseudo-Callisthenes] texts’.60
Besides the different hypotheses on the translation of the Syriac Pseudo-Callisthenes, whether it came through a Middle Persian Vorlage or was based on a Greek original, we have to consider the analysis of another branch of Alexander Romance tradition. As very few texts of pre-Islamic Persian literature have survived, the only evidence we have for the existence of such a Middle Persian Vorlage is Arabic and Persian sources that mention different motifs from the Pseudo-Callisthenes tradition.
Here another problem arises: ‘the Arabic PC [Pseudo-Callisthenes], if indeed it had ever existed, had been lost.’61 However, it is believed that such an Arabic version did exist, because various elements from the Pseudo-Callisthenes tradition can be found in historical and geographical works in Arabic.62 Furthermore, it is also important to point out the hypothesis regarding the development of the Pseudo-Callisthenes tradition in Arabic. According to Doufikar-Aerts, this hypothesis is generally accepted:
It is now clear that three Syriac texts dating from the early seventh century were the sources for a great amount of Arabic material. In the seventh century, there was a Syriac translation of the Greek δ-recension of Pseudo-Callisthenes. The Romance was mediated by Middle Persian, according to Nöldeke, and it acquired a veneer of Iranian names during this stage. The two other texts were original compositions in Syriac, written between 628 and 640, and they reflect the political ideologies of their period. It is from these latter texts that the Arabic Alexander derives his character as a prophet. An Arabic translation, based on this Syriac Pseudo-Callisthenes, was probably made in the ninth century. Then, between the fourteenth and sixteenth centuries, an adaptation of this Alexander Romance was made in Ethiopic, based on an Arabic Vorlage. Because of the similarity between the Ethiopic Romance and the Syriac Pseudo-Callisthenes, the Arabic translation is generally believed to have been the intermediary between the Syriac and the Ethiopic Romance. This is just a hypothesis, since the Arabic translation has never been found and is thought to have been lost.63
This book will address two different areas of this debate. Firstly, the scholars mentioned above (Nöldeke, Ciancaglini, van Bladel and Monferrer-Sala) based their hypotheses on a philological examination of the Syriac Pseudo-Callisthenes. This causes some problems due to the date and origin of the Syriac manuscripts, the oldest of which dates only to 1709.64 The influence of scribes, and the possible elements they might have added to the manuscripts, causes confusion when it comes to a textual and philological examination. In order to solve this problem, we should leave aside the philological examination of the Syriac text and instead explore its content in parallel with the Arabic and Persian texts. It is thus highly important to focus on the significance of the Sasanian legacy in these sources.
Secondly, an examination of the extensive Arabic tradition on Alexander shows two different branches: those versions that presented Alexander as a Persian king of the Kayānid dynasty, by naming him the half-brother of Darius III (Dārā), and the versions that did not. Doufikar-Aerts's excellent study on the different branches of the Arabic Alexander tradition missed an important element, namely the legacy of the Sasanian sources, especially in Arabic historiography. Due to this omission, she did not offer any solution to the problem of the ‘lost Middle Persian Vorlage’ in the transmission of the Pseudo-Callisthenes in the Arabic tradition. This book will examine the content of relevant Arabic and Persian sources in which it is possible to trace the Pseudo-Callisthenes tradition in a new light; this will show the influence of Sasanian sources on Arabic and Persian literature, especially in the field of historiography.
The Greek Alexander Romance originated in Alexandria (Egypt) in the third century AD and became the most influential source for the deeds and adventures of Alexander, especially in the East. It is therefore important to study its influence in the Persian versions of the tale. However, there is a huge gap between its composition in Alexandria in the third century AD and its appearance in the Persian tradition represented by the Shāhnāma of Firdawsī (completed in 1010), which is the first complete exemplar of the Pseudo-Callisthenes tradition. In order to fill this gap of seven centuries and provide a hypothesis regarding how this Greek work appeared in the Persian sources, we need to first consider the historical environment in which national Persian history was formed.