3.1 The List of Ḥamza

To understand the translation history of the Khwadāynāmag, we have to start with the best informed of all later authors, Ḥamza al-Iṣfahānī, Ta⁠ʾrīkh, pp. 9–10:¹

Their (the Persians’) chronologies are all confused, rather than accurate, because they have been transmitted for 150 years² from one language into another and from one script, in which the number signs are equivocal, into another language, in which the “knotted” number signs are also equivocal. In this chapter, I have had to take the recourse of collecting variously transmitted manuscripts (nusakh),³ of which I have come across eight, namely:

• H1. Kitāb Siyar mulūk al-Furs, translated/transmitted (min naql)⁴ by Ibn al-Muqaffaʿ;

• H2. Kitāb Siyar mulūk al-Furs, translated/transmitted (min naql) by Muḥammad ibn al-Jahm al-Barmakī;

• H3. Kitāb Ta⁠ʾrīkh mulūk al-Furs, which was taken from the Treasury (i.e., the Caliphal library) of al-Maʾmūn;

• H4. Kitāb Siyar mulūk al-Furs, translated/transmitted (min naql) by Zādūye ibn Shāhūye al-Iṣbahānī;

• H5. Kitāb Siyar mulūk al-Furs, translated/transmitted (min naql) or compiled (aw jamʿ) by Muḥammad ibn Bahrām ibn Miṭyār al-Iṣbahānī;

• H6. Kitāb Ta⁠ʾrīkh mulūk Banī Sāsān, translated/transmitted (min naql) or compiled (aw jamʿ) by Hishām ibn Qāsim al-Iṣbahānī;

• H7. Kitāb Ta⁠ʾrīkh mulūk Banī Sāsān, corrected (min iṣlāḥ) by Bahrām ibn Mardānshāh,⁵ the mōbad of Kūrat Sābūr of the province of Fārs.

When I had collected them I compared them with each other until I managed to compile what is correct in this chapter.

As will later be shown, the missing eighth author is Mūsā ibn ʿĪsā al-Kisrawī.

It has usually been taken for granted that all the seven books mentioned on this list were translations of the Khwadāynāmag, but Ḥamza himself does not claim that this is so. He is merely speaking about the chronology of pre-Islamic Persian kings and about manuscripts which contained information on them, without specifying whether he is speaking of copies of one original work or of several different works. As some of the works are implied to have only been concerned with the Sasanians and as the Khwadāynāmag seems to have taken up the story from Gayōmard onward, it seems extremely improbable that all books on the list were translations of the Khwadāynāmag.

Ḥamza himself was not a translator and no translations from Middle Persian are attributed to him (see also Chapter 3.6). Moreover, he clearly speaks of translations from one language into another, which shows that the listed texts were in Arabic. Thus, one has to take H7 as just what it is said to be, namely “Kitāb Ta⁠ʾrīkh mulūk Banī Sāsān, corrected (min iṣlāḥ) by Bahrām ibn Mardānshāh, the mōbad of Kūrat Sābūr of the province of Fārs,” i.e., an Arabic book corrected by a Zoroastrian scholar; likewise, H4, Zādūye ibn Shāhūye bears a non-Islamic name but writes in Arabic. Whether the corrections of Bahrām were based on some Middle Persian manuscript(s) (perhaps, but not necessarily, the Khwadāynāmag), his own general knowledge of Persian national history, or some other Arabic texts, such as variant versions/translations of the Khwadāynāmag, is not stated and should not without further study be claimed in one way or the other.

Ḥamza’s list may be compared with that of al-Bīrūnī, Āthār, p. 114/99//107–108:

This is according to what I have heard from Abū l-Ḥasan Ādharkhwar the Architect (al-Muhandis). Abū ʿAlī Muḥammad ibn Aḥmad al-Balkhī al-Shāʿir⁶ has told in al-Shāhnāme the story of the origin of mankind differently from what we have narrated. He claims that he revised his report on the basis of:

• B1. the Kitāb Siyar al-mulūk which is by ʿAbdallāh Ibn al-Muqaffaʿ [H1];

• B2. and the one by Muḥammad ibn al-Jahm al-Barmakī [H2];

• B3. and the one by Hishām ibn al-Qāsim [H6];

• B4. and the one by Bahrām ibn Mardānshāh, the mōbad of the city of Sābūr [H7];

• B5. and the one by Bahrām ibn Mihrān al-Iṣbahānī [= H5?].

These he collated with what

• B6. Bahrām al-Harawī al-Majūsī brought him.⁷

Abū ʿAlī Muḥammad al-Balkhī (as quoted by al-Bīrūnī), thus, omits the anonymous al-Ma⁠ʾmūn manuscript and the Zādūye version and, like the preserved manuscript of Ḥamza’s Ta⁠ʾrīkh, does not mention Mūsā ibn ʿĪsā al-Kisrawī.

Both lists may further be compared with Ibn al-Nadīm’s list of Persian translators in the Fihrist, p. 305/245//589. Ibn al-Nadīm’s list is somewhat confused and has never been properly discussed. The subchapter is entitled “The Names of the Translators (al-naqala)⁸ from Persian into Arabic” and it begins with the mention of Ibn al-Muqaffaʿ and others who have just been discussed by Ibn al-Nadīm and who do not seem to have been specifically or solely working with the Khwadāynāmag. The list ends with Isḥāq ibn Yazīd (see Chapter 3.2.7).

After this the text continues: wa-min naqalat al-Furs, followed by a list with mostly the same names that are on Ḥamza’s list – the absence of Ibn al-Muqaffaʿ is explicable by his having been mentioned a couple of lines earlier. The names listed are:

• N1. Muḥammad ibn al-Jahm al-Barmakī [H2];

• N2. Hishām ibn al-Qāsim [H6];

• N3. Mūsā ibn ʿĪsā al-*Kisrawī;⁹

• N4. Zādūye ibn Shāhūye al-Iṣbahānī [H4];

• N5. Muḥammad ibn Bahrām ibn Miṭyār al-Iṣbahānī [H5];

• N6. Bahrām ibn Mardānshāh, the mōbad of the city of Sābūr [H7];

• N7. ʿUmar ibn al-Farrukhān.¹⁰

There can be little doubt that Ibn al-Nadīm is here dependent on some source or sources that belong to the same tradition as that used by Ḥamza, or on Ḥamza himself, even though he does not mention Ḥamza by name.¹¹ The only additional names are Mūsā ibn ʿĪsā, erroneously dropped from Ḥamza’s list (cf. below), and ʿUmar ibn al-Farrukhān (on whom, see Chapter 3.2.8).

There are still three further sources to be considered. The anonymous Persian Mujmal al-tawārīkh mentions (p. 2/2) among its sources the collection of Ḥamza (majmūʿe-ye Ḥamza ibn al-Ḥasan al-Iṣfahānī), who transmitted from the works of:

• M1. Muḥammad ibn al-Jahm al-Barmakī [H2];

• M2. Zādūye ibn Shāhūye al-Iṣfahānī [H4];

• M3. Muḥammad ibn Bahrām ibn Miṭyān/r [H5];

• M4. Hishām ibn al-Qāsim [H6];

• M5. Mūsā ibn ʿĪsā al-Kisrawī;¹²

• M6. o-kitāb-e tārīkh-e pādishāhān iṣlāḥ-e Bahrām ibn Mardānshāh mōbad-e Shāpūr az shahr-e Pārs bīrūn āwurde-ast. [H7]

The list admittedly depends on Ḥamza. The lack of H1, Ibn al-Muqaffaʿ, is again explicable by his having been mentioned immediately before Ḥamza, the repetition being avoided by dropping the name from Ḥamza’s list. Further, the al-Ma⁠ʾmūn manuscript (H3) is dropped, which may be a simple mistake. The last words of M6 come curiously close to Ḥamza’s description of the manuscript taken (al-mustakhraj) from al-Ma⁠ʾmūn’s Treasury (H3). Note that in the Older Preface (cf. below) al-Ma⁠ʾmūn’s manuscript and Bahrām ibn Mardānshāh follow each other (OP9–OP10), which makes it possible that the list of the Mujmal is corrupt and the al-Ma⁠ʾmūn manuscript has been dropped by mistake, which would make the last words an attempt to make sense of the corrupt passage. Hence, M6 may hide behind itself two different books, the manuscript of the History of the Kings of Persia (*Ta⁠ʾrīkh mulūk al-Furs) taken from (cf. bīrūn āwurde ast) al-Ma⁠ʾmūn’s Treasury and Bahrām ibn Mardānshāh’s book on Sasanian kings. The addition of Mūsā ibn ʿĪsā will be discussed below.

The fifth list is found in Balʿamī’s Tārīkhnāme i: 5.¹³ The list is partly confused. Balʿamī quotes the following as his authorities:

• BL1. Shāhnāme-ye buzurg-e Ḥamza-ye Iṣfahānī;¹⁴

• BL2. pisar-e Muqaffaʿ yaʿnī ʿAbdallāh [H1];

• BL3. Muḥammad ibn al-Jahm al-Barmakī [H2];

• BL4. Zādūye ibn Shāhūye [H4];

• BL5. nāme-ye Bahrām ibn Bahrām [= H5?];

• BL6. nāme-ye Sāsāniyān;

• BL7. Mūsā ibn ʿĪsā al-Khusrawī;¹⁵

• BL8. Hāshim o-Qāsim-e [sic] Iṣfahānī¹⁶ [H6];

• BL9. pādishāhān-e Pārs;

• BL10. (Zādūy-e)¹⁷ Farrukhān mōbad-e mōbadān.¹⁸

The sixth and final list is that given in the Older Preface to the Prose Shāhnāme. The text of this list is slightly confused. My readings are explained in Chapter 7.4:¹⁹

• OP1. nāme-ye pisar-e Muqaffaʿ [H1];

• OP2. (nāme-ye) Ḥamza-ye Iṣfahānī;

• OP3. Muḥammad-e Jahm-e Barmakī [H2];

• OP4. Zādūy ibn Shāhūy [H4];

• OP5. nāme-ye Bahrām-e [Mihrān-e] Iṣfahānī [= H5?];

• OP6. nāme-ye Sāsāniyān-e Mūsā-ye ʿĪsā-ye Khusrawī;

• OP7. Hishām-e Qāsim-e Iṣfahānī [H6];

• OP8. nāme-ye shāhān-e Pārs;

• OP9. az ganj-khāne-ye Ma’mūn²⁰ [H3];

• OP10. Bahrāmshāh-e Mardānshāh-e Kirmānī [H7];

• OP11. Farrukhān, mōbadhān mōbadh-e Yazdagird-e Shahriyār;

• OP12. Rāmīn ke bande-ye Yazdagird-e Shahriyār būd.

We may now compare the six lists with each other:

View Table

The table is rather clear. Ibn al-Muqaffaʿ’s absence from Ibn al-Nadīm’s Fihrist and the Mujmal is easily explicable, as he has been mentioned a few lines earlier in both sources and his absence from this list merely avoids repetition. The anonymous manuscript “from the Treasury of al-Ma⁠ʾmūn” seems to have fallen victim of scribal errors in several sources, cf. above.

If we equate Abū ʿAlī al-Balkhī’s Bahrām ibn Mihrān with Ḥamza’s Muḥammad ibn Bahrām ibn Miṭyār, or consider him Muḥammad’s father, then Zādūye’s absence from al-Balkhī’s list is probably accidental as it would seem that al-Balkhī has otherwise merely copied the list from Ḥamza, possibly from a manuscript from which Mūsā’s name had already been dropped. On the other hand, the resemblance of the two lists might itself be accidental, in which case Zādūye’s absence from the list merely means that he was not used by Abū ʿAlī al-Balkhī, who really used, or at least had seen, the other sources he mentioned. However, I am ready to opt for the first explanation. In that case al-Balkhī’s seemingly impressive list turns out to have been copied from Ḥamza.

As Ḥamza wrote around the mid-tenth century and Abū ʿAlī al-Balkhī’s date is not known (cf. Chapter 4.1.2), it might also be possible to turn the tables and claim that it was Ḥamza who lifted the list from al-Balkhī.²¹ In this case we should also assume that al-Bīrūnī, or his informant, for some reason dropped the titles of several books on the list, which is not very probable and tips the balance in Ḥamza’s favour. In both cases, however, it should be noted that al-Balkhī was using Arabic sources, either Ḥamza (from whom he lifted the whole list) or a series of Arabic authors (certain for B1 and B2, probable because of the Islamic name and patronym in the case of B3, and possible or probable in the remaining two cases).

For ʿUmar ibn al-Farrukhān and Rāmīn, see Chapters 3.2.8 and 3.2.10.

Ḥamza lacks Mūsā ibn ʿĪsā, from whom he quotes soon after in extenso (pp. 16–21). Al-Kisrawī’s book can hardly be equated with the anonymous manuscript from al-Ma⁠ʾmūn’s Treasury, as the Older Preface gives on its list both and as most sources would indicate that Mūsā ibn ʿĪsā lived somewhat later (Chapter 3.3).

Ḥamza claims to be listing eight sources while actually naming only seven.²² The above table shows clearly that Mūsā’s book has been accidentally dropped from Ḥamza’s list. Comparing the order of the items listed in the various sources, we may surmise that Mūsā was either listed before Hishām (Balʿamī, the Older Preface) or after him (the Mujmal, Ibn al-Nadīm).

The analysis of these lists has an important consequence for the question of the Arabic translations of the Khwadāynāmag.²³ There is no specific reason to doubt Ḥamza’s, or the other authors’, reliability, yet one cannot refrain from noting that the list of eight names (H1–7 + Mūsā) is repeated from one source to the other, mainly in the same order and with few changes or additions, which makes one doubt whether the authors who listed them really had used, or even seen, them, or whether they just lifted the list from an earlier source to include it in their own book to show off their meticulous scholarship, much like a modern scholar would lift an impressive list of scholarly references from an earlier study without actually having read them.²⁴ It seems that we only have Ḥamza’s word for the existence of some of these translations or reworkings.

3.2 Translators and Their Translations

This chapter briefly studies the authors mentioned in Chapter 3.1 and adds two further informants in Chapter 3.2.11. Logically, this chapter should begin with the first known translator, Ibn al-Muqaffaʿ, but as he is a special case because of the relatively large amount of information we have both on him and his translation, he will be discussed in a separate chapter (3.4). Mūsā ibn ʿĪsā al-Kisrawī, or Khusrawī, will also be dedicated a separate chapter (3.3) for similar reasons.

3.3 Mūsā ibn ʿĪsā al-Kisrawī

In tenth-century sources, a Mūsā ibn ʿĪsā al-Kisrawī, or Khusrawī, is sometimes referred to, but we know little about his life and activities.⁵⁸ The aim of this chapter is to discuss the scant evidence at our disposal and to shed at least some light on this shadowy character, even though in the end we still have to admit that we know little about who he was and what he did.

Mūsā ibn ʿĪsā was on the original list of Ḥamza (Chapter 3.1), though his name was rather early dropped from it. If we take nāme-ye Sāsāniyān to be the title of Mūsā ibn ʿĪsā al-Kisrawī’s book in the Preface of the Prose Shāhnāme and in Tārīkh-e Balʿamī – it would fit the supposed contents of the book, cf. below – the missing of Mūsā’s name from Ḥamza’s list could be explained as a copyist’s error. For the original “*Kitāb Ta⁠ʾrīkh mulūk Banī Sāsān (nāme-ye Sāsāniyān in the Persian translation) by al-Kisrawī and xxx by Hishām” the copyist inadvertently dropped al-Kisrawī’s name and the following title, thus reducing the number of authors from eight to seven. In the Mujmal and the Fihrist, though, it should be emphasized, Hishām comes before Mūsā, not after him, which makes this explanation problematic. Thus, we cannot be sure whether *Kitāb Ta⁠⁠ʾrīkh mulūk Banī Sāsān was the title of his book.

Mūsā ibn ʿĪsā al-Kisrawī is firmly established on the list in several sources, though accidentally dropped from the original. But what was his book like?⁵⁹ The term naql, used in Ḥamza’s list, is ambivalent and Mūsā ibn ʿĪsā hardly “translated” anything, at least for this work, but more probably wrote a Persian history based on some original historical source(s) translated from Pahlavi. Mūsā may have synchronized Persian history with the sacred history or he may also have written a rather dry chronology, as far as we can deduce from Ḥamza’s Ta⁠ʾrīkh (for other sources, see below). Whether Mūsā was able to use Middle Persian texts in the original language is questionable. At least in the long quotation in Ḥamza’s Ta⁠ʾrīkh he is speaking of Arabic translations (cf. below).⁶⁰

The possible contents of this lost book may now be discussed in the light of the admittedly rather sparse evidence.

In Ḥamza’s Ta⁠ʾrīkh, pp. 16–21, there is a long quotation from, or perhaps partly a paraphrase of, al-Kisrawī’s book. This is our most reliable and the only unproblematic piece of evidence as to the contents and date of this lost book. However, one has to remember that Ḥamza himself was mainly interested in chronology and his selection may, thus, give a distorted picture of what his sources really contained. But at least we know that, perhaps among other materials, Mūsā’s work contained chronological information. The beginning of this passage deserves to be translated in toto:

Mūsā ibn ʿĪsā al-Kisrawī has said in his book: I looked into the book called the Khudāynāme, which is the book that, when translated from Persian into Arabic, is called Ta⁠ʾrīkh⁶¹ mulūk al-Furs. I repeatedly looked into manuscripts (nusakh) of this book and perused them minutely, finding that they differ from each other. I was unable to find two identical copies. This is because the matter had been confused by the translators of this book when they translated it from one language into another. When I was together with al-Ḥasan ibn ʿAlī al-Hamadānī al-Raqqām in Marāgha at (the court) of its ruler (ra⁠ʾīs) al-ʿAlāʾ ibn Aḥmad … (the text continues to tell how they collated the overall lengths of the third and fourth dynasties with the Alexandrian era as found in astronomical tables).⁶²

The sentence “This is because the matter had been confused by the translators of this book when they translated it from one language into another” is crucial as it shows that Mūsā worked with translations, not versions of the original Middle Persian text.⁶³ Whether he knew Middle Persian or not cannot be deduced from this or any other passage.

At the end of the passage quoted from Mūsā ibn ʿĪsā (pp. 20–21), there is an important note on the chronology of the pre-Sasanian Kings. Whereas al-Kisrawī seems very proud of his accuracy when it comes to Sasanian history,⁶⁴ he admits that he did not study the earlier period in such detail, claiming that Alexander’s misdemeanour in Iran had disrupted the tradition so that no accuracy in earlier chronology is possible:⁶⁵

I have not concerned myself with the chronologies of the Ashghānian kings before the Sasanians because of the misfortunes that occurred at the time of those kings. Namely, when he had conquered the land of Babel, Alexander envied the sciences that they (i.e., the Persians) had acquired, such as no nation had been able to acquire before. He burned all their books he was able to find and then turned to killing their mōbads and hērbads and learned and wise men and those who, among their other sciences, preserved their chronologies, until he had killed them all. This he did after he had translated (naqala) what he needed of their sciences into Greek.⁶⁶ After this, during all the days of the Ashghānians, also known as the Petty Kings, the Persians remained obscure (ghāba), having no one to bring back knowledge or to be concerned with any kind of wisdom until their rule (dawla) returned to them with the appearance of Ardashīr.

When Ardashīr confirmed the kingship for himself, he started counting time from his own accession. After him, the Sasanian kings followed his way and each of them counted time by his own regnal years, which has caused confusion in their chronologies. What an excellent idea it was that the Arab kings decided to count their years continuously, from the beginning of the hijra onward.

The passage implies that al-Kisrawī may not, except in broad outlines, have discussed this period at all, at least not in chronological terms. It would be somewhat strange to see an author first undermine his own authority and then delve into this period. Possibly, the book of al-Kisrawī was restricted to the Sasanian period only, which would speak for taking *Kitāb Ta⁠ʾrīkh mulūk Banī Sāsān as its title.⁶⁷ This would also mean that he did not translate the Khwadāynāmag, but merely used its Sasanian part as an authoritative source for his own book.

In the rest of his work, Ḥamza is unfortunately vague in identifying his sources, usually using expressions such as kutub al-siyar, baʿḍ al-ruwāt, zaʿamat al-Furs, wa-fī akhbārihim, etc.⁶⁸ Thus, we cannot know whether he used any other parts of al-Kisrawī’s book or, in fact, whether al-Kisrawī’s book was merely a chronological list. In the quotation from al-Kisrawī, Ḥamza, Ta⁠ʾrīkh, p. 20, mentions Būrāndukht bint Kisrā Abarwīz, saying that it was she who returned the True Cross (wa-hiya allatī raddat khashabat al-Masīḥ). The interest in Christian history makes it improbable that this could be a direct quotation from any Middle Persian, pre-Islamic source, such as the Khwadāynāmag, so that we may assume that al-Kisrawī added notes and comments to the text he was working with or that these additions were already made in the text(s) he used.⁶⁹

Ḥamza’s Ta⁠ʾrīkh provides us with our only unproblematic and reliable source of information on Mūsā ibn ʿĪsā’s book and its contents. An “al-Kisrawī” is also mentioned or quoted in a number of other sources, but rarely identified more exactly, and his identity remains uncertain, as there is also another al-Kisrawī, ʿAlī ibn Mahdī, who at least in some cases may be the person referred to.

Ps.-al-Jāḥiẓ, Maḥāsin, quotes al-Kisrawī – always without a first name – three times (pp. 53, 242, 359). The first passage (p. 53, from al-Bayhaqī, Maḥāsin, p. 534) concerns Sasanian history, being a brief saying by Kisrā ibn Hurmuz, and the second (pp. 242–251) is a long romantic story about the Indian marriage of the Parthian Balāsh ibn Fīrūz, containing two framed animal stories, material that had little place in the Khwadāynāmag of the Sasanians.

Balāsh usually receives scant interest in historical sources.⁷⁰ An important exception is the anonymous Nihāyat al-arab,⁷¹ which seems to be where al-Kisrawī took this story from (pp. 277/280–294), and then either he or the anonymous author of the Maḥāsin abbreviated it.⁷² The story is also referred to in Mujmal, p. 58/72,⁷³ where the anonymous author mentions that he had read it in Siyar al-mulūk (dar Siyar al-mulūk khwāndam). As the al-Kisrawī quotations in the Maḥāsin and the Nihāya are the only preserved versions of this story, the passage should be given due attention. Usually, the quotations from Siyar al-mulūk in the Mujmal and in other sources are all too hastily taken as quotations from Ibn al-Muqaffaʿ’s work. This, however, is ungrounded and each quotation should be studied separately. It is, of course, possible that Ibn al-Muqaffaʿ’s influential text contained this story, but in that case one might wonder why it was taken up by so few later sources. A less-known al-Kisrawī would understandably be quoted by only a few. On the other hand, it should be emphasized that when al-Kisrawī is quoted by name (and translated into Persian) in the Mujmal, this is always done through Ḥamza (pp. 2/2, 67/85,⁷⁴ 68/87, 70/88). Hence, there is no evidence to show that the author of the Mujmal would have had al-Kisrawī’s book to hand.

It is difficult to contextualize the Balāsh story. Though set in a historical context, it differs from the tone of the other early sources that derive material from the Khwadāynāmag, whether in Persian or Arabic, which contain no framed stories, animal or otherwise, and give more emphasis to the epic-heroic than to the romantic material, and we have to come up to Firdawsī before finding similar material, and even there framed narratives are rare.⁷⁵ Hence, it remains doubtful whether the passage could stem from any translation/rewriting of the Khwadāynāmag. Al-Kisrawī’s book may, of course, have been far from the main stream of the tradition and contained more novelistic and romantic material than many other representatives of the tradition, as suggested by Rozen (1895), but it should be emphasized that his hypothesis rests solely on the identification of al-Kisrawī in this passage with Mūsā ibn ʿĪsā, which is far from evident.⁷⁶ If the passage comes from Mūsā ibn ʿĪsā’s book, it would still say nothing about the Khwadāynāmag and its Arabic translations, as there is no reason to assume that Mūsā ibn ʿĪsā could not have used other sources, too, and the overwhelming majority of evidence points to the rather dry character of the Khwadāynāmag and its translations.

The final passage transmitted from al-Kisrawī in ps.-al-Jāḥiẓ, Maḥāsin, comes in the Chapter entitled Maḥāsin al-nayrūz wa’l-mihrajān (p. 359ff.) and probably continues until p. 365.⁷⁷ It is concerned with the nawrūz (= nayrūz). The passage contains an important description of the ceremonies of the nawrūz and the mihrajān, mentioning also songs, some of them obviously epic, which were sung in the presence of the King.⁷⁸

This passage might well come from the Book of Festivals, Kitāb al-aʿyād wa’l-nawārīz, attributed to ʿAlī ibn Mahdī al-Kisrawī (cf. below). As it is somewhat uneconomic to suggest that the anonymous author of the Maḥāsin derived material from two different al-Kisrawīs,⁷⁹ one should consider the possibility that all quotations come from the same al-Kisrawī. The first quotation could well be from Mūsā’s book and the second, too, is not inconceivable as part of his book, even though the part preserved by Ḥamza consists of a rather dry chronology and the early fragments attributed to Ibn al-Muqaffaʿ do not include very much romance.

The third passage is the most difficult to fit into Mūsā’s work. The establishment of nawrūz and of mihrajān quite centrally belong to Persian national history, but later rituals do not. ʿAlī ibn Mahdī’s book, on the other hand, would be an excellent place for this third fragment and the second would fit another book of his, Kitāb al-Khiṣāl (see below), as would the first. Attributing all three passages to ʿAlī ibn Mahdī may be easier than attributing all of them to Mūsā ibn ʿĪsā,⁸⁰ although the problem remains that we should posit two separate books as the sources for the three quotations. There is also a further problem. Ibn Isfandiyār’s Tārīkh Ṭabaristān, for which see below, again confuses the picture by giving us some ground for asking whether the Book of Festivals was, after all, by ʿAlī ibn Mahdī or whether it could have been authored by Mūsā ibn ʿĪsā.

Much of the material in this third quotation is unique, even though, in general terms, e.g., al-Bīrūnī, Āthār, pp. 263–289/215–233//199–219, and Gardīzī, Zayn, pp. 345–355, resemble it in their descriptions of these festivals, but the resemblance may well be merely due to the common object of description and not evidence of any textual dependence. The verse by Abū Tammām, quoted in the Maḥāsin, p. 360, is commonly found in the historical tradition that is dependent on al-Ṭabarī’s Ta⁠ʾrīkh, but in the Maḥāsin there is an interesting variant in the first hemistich (wa-ka’annahū l-Ḍaḥḥāku fī fatakātihī), whereas all other sources have the standard version (bal kāna ka’l-Ḍaḥḥāki fī saṭawātihī), which is also the Dīwān recension.⁸¹ This seems to point to an independent line of transmission, even though one cannot exclude the possibility of later manuscript corruption.

Al-Bīrūnī, Āthār, contains three quotations from al-Kisrawī (pp. 135, 144–146, 273/119, 129–131, 223//122, 127–128, 208). The first two are explicitly taken from Ḥamza (Āthār, p. 135: wa-ammā Ḥamza al-Iṣfahānī fa-innahu ḥakā ʿan Mūsā ibn ʿĪsā) and paraphrase, condense, and criticize Ta⁠ʾrīkh, pp. 16–21.

However, the third passage (p. 273/223//208)⁸² mentions a new character (on whom, see Chapter 3.2.11): wa-qāla l-Kisrawī: samiʿtu al-mōbad al-Mutawakkilī yaqūlu. This passage is not found in Ḥamza’s Ta⁠ʾrīkh, which shows that this book was not the sole source for al-Bīrūnī’s al-Kisrawī material.⁸³ The passage concerns the mihrajān and, likewise, is unattested elsewhere. This passage might equally well come from ʿAlī ibn Mahdī al-Kisrawī’s Book of Festivals, as the personal name of al-Kisrawī is not indicated. In any case, the third quotation comes from another source than Ḥamza’s Ta⁠ʾrīkh, which is the source for the first two quotations.

Finally, there is an interesting passage in Ibn Isfandiyār’s Tārīkh Ṭabaristān (written in 616/1216), p. 83, which gives us reason to reconsider the authorship of the Book of Festivals:

In order not to be attacked by the readers claiming that I have lied I have left out the stories about Bīwarasb and what happened to him, which the Caliph Ma⁠ʾmūn ʿAbdallāh ordered to be enquired into,⁸⁴ and (what happened) during the reigns of Hurmizd-shāh and Khusraw Parwīz and the story of Mūsā ibn ʿĪsā al-srwy (read: al-Kisrawī),⁸⁵ which is related in the book Nayrūz wa-mihrajān, and the story of the Slavegirl and Ḥurra al-Yasaʿiyya because they are far from reason and are not among the stories of the people of the Sharīʿa.

The otherwise unknown “story of Mūsā ibn ʿĪsā al-Kisrawī” should probably be understood as a story (related) by Mūsā ibn ʿĪsā, not a story about him. This would still be our only source attributing this text to Mūsā ibn ʿĪsā, whereas all other sources attribute it to ʿAlī ibn Mahdī.

On this basis, we may now sketch the contents of al-Kisrawī’s book. Two things highlight themselves. The material that we can certainly attribute to Mūsā ibn ʿĪsā al-Kisrawī is the dry chronological data on the Sasanids in Ḥamza’s Ta⁠ʾrīkh, which tallies well with the speculation concerning the book’s title.

In addition, a certain al-Kisrawī, either Mūsā ibn ʿĪsā or ʿAlī ibn Mahdī, is credited with long narratives, some of which make use of framed stories, a feature we can find nowhere else in the sources that should contain material derived from the Khwadāynāmag. If they derive from a work by ʿAlī ibn Mahdī they have nothing to do with the Khwadāynāmag, as ʿAlī ibn Mahdī is nowhere attached to the Khwadāynāmag or its Arabic translations.

If the stories come from Mūsā ibn ʿĪsā’s book, then, as Rozen has already pointed out,⁸⁶ it is dubious whether we can properly call this book a translation of the Khwadāynāmag. Al-Kisrawī would have made substantial additions to his text and, if the Khwadāynāmag started from the Creation, as seems probable, may even have deleted a major portion of the original. In short, it may be more to the point to take his work as a new book, partly based on the materials in the Khwadāynāmag (in Arabic translation).

Finally, we come to the question of Mūsā ibn ʿĪsā’s identity. The long quotation from him in Ḥamza, Ta⁠ʾrīkh, pp. 16–21, provides us with the basic facts of his life. He collaborated with al-Ḥasan ibn ʿAlī al-Hamadhānī al-Raqqām in Marāgha, when the town was under al-ʿAlāʾ ibn Aḥmad. They collated various chronologies, using Zīj al-raṣad, to create a more reliable chronology of Persian history. Mūsā also quotes (Kitāb) al-Siyar al-kabīr and (Kitāb) al-Siyar al-ṣaghīr (Ta⁠ʾrīkh, p. 20), which shows that he depended on at least two different redactions of Persian national history in Arabic translation.⁸⁷

Al-ʿAlāʾ ibn Aḥmad al-Azdī’s governorship of Marāgha gives us some firm ground for dating Mūsā. Al-ʿAlāʾ died in 260/874 when Governor of Ādharbayjān.⁸⁸ This would date Mūsā’s activity with Sasanian chronology probably in the 860s or early 870s. If he is the al-Kisrawī who transmitted from al-mōbad al-Mutawakkilī, this would, for its part, confirm Mūsā’s date around 870.

The Fihrist’s list of translators/transmitters of Persian books has already been discussed (Chapters 3.1 and 3.2.7), but Ibn al-Nadīm also knew two other books by Mūsā ibn ʿĪsā (Fihrist, p. 142/128//280), neither of which presumably contained specifically Persian material, namely:

• Kitāb Ḥubb al-awṭān

• Kitāb Munāqaḍāt man zaʿama annahu lā yanbaghī an yaqtaḍiya l-quḍāt fī maṭāʿimihim bi’l-a⁠ʾimma wa’l-khulafāʾ⁸⁹

He is also credited with these two books in Ismāʿīl Pāshā’s Hadiyyat al-ʿārifīn, p. 477, where we have some additional pieces of information. First, he is called Mūsā ibn ʿĪsā al-Baghdādī⁹⁰ al-adīb al-shahīr bi’l-Kisrawī and, secondly, he is said to have died in 186, which is too early a date in comparison with all the other evidence. We might consider an emendation to *286, though it remains unclear where Ismāʿīl Pāshā got the date from.

Besides knowing his al-Ḥanīn ilā l-awṭān (sic, gal S i: 945, sub 237)⁹¹ Brockelmann credits Mūsā ibn ʿĪsā with a translation, or version, of Sindbādnāme (gal S i: 237), but this seems to be a wild guess with little real foundation.⁹²

Yāqūt, cf. below, at one point refers to Mūsā ibn ʿĪsā as al-Kisrawī al-Kātib. This is the only indication that he was a kātib, but as many of the translators from Persian as well as transmitters of Persian lore worked as government officials, this would not, a priori, be surprising. However, there is a possibility of confusion here, as al-Kisrawī al-Kātib would usually seem to refer to ʿAlī ibn Mahdī.

This, nevertheless, gives us some room for speculation. In his Wuzarāʾ, p. 407, al-Jahshiyārī mentions an otherwise unknown Mūsā ibn ʿĪsā ibn YazdānYRWDh, who was a scribe working for al-Faḍl ibn al-Rabīʿ (kāna yaktubu li’l-Faḍl ibn al-Rabīʿ) during the Caliphate of al-Amīn.⁹³ It is not impossible that this scribe should be identified with our al-Kisrawī. His name proves that he was of Persian extraction, as we would suppose al-Kisrawī to have been, and like most translators from Middle Persian and transmitters of Persian lore were. Further, he worked as a scribe and we have every reason to believe, whether al-Kisrawī al-Kātib refers to him or not, that Mūsā ibn ʿĪsā came from the same circles, as did most of the others who worked within the Book of Kings tradition. Dating him to the period of al-Amīn (and supposing him to have lived on several decades after al-Amīn’s death) tallies well with the known interest in Persian national history during the early to mid-ninth century (and later). This identification would also count for the gentilicium al-Baghdādī given to him in Ismāʿīl Pāshā’s Hadiyya, though one should not put too much weight on this rather suspect piece of information. Hence, the least we can say is that there is nothing to preclude this identification. On the other hand, of course, there is no positive evidence that Mūsā ibn ʿĪsā al-Kisrawī was the grandson of a certain YazdānYRWDh, and there is a slight temporal gap between the two. Hence, the identification remains highly speculative.⁹⁴

This more or less sums up what we know about Mūsā ibn ʿĪsā al-Kisrawi. The other al-Kisrawī, ʿAlī ibn Mahdī, is also credited with one of the books attributed to his namesake, Mūsā ibn ʿĪsā, namely Kitāb Munāqaḍāt, even in the very same source (Fihrist, p. 167/150//328). This shows how confused tenth-century authors were about the identity of al-Kisrawī.

ʿAlī ibn Mahdī is also credited in the same passage of the Fihrist with a Kitāb al-aʿyād wa’l-nawārīz, which is not extant, but the title would imply that it contained material about the Nawrūz and, most probably, the Mihrajān, i.e., the very kind of material which we have often seen transmitted on the authority of al-Kisrawī. As we have seen, though, Ibn Isfandiyār, Tārīkh Ṭabaristān, may attribute this book to Mūsā ibn ʿĪsā, but it is the only source to do so. Interestingly enough, al-Bīrūnī, Āthār, 38/31//36, mentions a tractate by Ḥamza al-Iṣfahānī on poems on the Nawrūz and the Mihrajān.

ʿAlī ibn Mahdī ibn ʿAlī ibn Mahdī al-Kisrawī Abū l-Ḥasan al-Iṣfahānī is mentioned in several biographical dictionaries. Yāqūt, Irshād iv: 334–338, has an article on him, saying, among other things, that he was the teacher of the son of Abū l-Ḥasan ʿAlī ibn Yaḥyā ibn al-Munajjim and aḥad al-ruwāt al-ʿulamāʾ al-naḥwiyyīn al-shuʿarāʾ at the time when Badr al-Muʿtaḍidī was the ruler of Isfahan (i.e., 283–289/896–902). Yāqūt seems to have (directly or indirectly) quoted from a work by Ḥamza (presumably his Ta⁠ʾrīkh Iṣfahān, which he also quotes by referring to the book title but without mentioning the author’s name in Irshād iv: 338) and explicitly says that al-Marzubānī mentioned him, quoting also Ibn Abī Ṭāhir. He also mentions his close association with Kitāb al-ʿAyn.⁹⁵

Yāqūt, Irshād iv: 336, specifically qualifies ʿAlī ibn Mahdī as aḥad al-ruwāt li’l-akhbār, but unfortunately does not, in the whole article, quote anything that would link him with any pre-Islamic Iranian material. ʿAlī ibn Mahdī’s date, however, is not too late for him to be the al-Kisrawī quoted in any of the sources discussed above. Yāqūt also mentions the following works by ʿAlī ibn Mahdī:

1. Kitāb al-Khiṣāl, a collection of stories (akhbār), wise sayings, proverbs, and poems.⁹⁶

2. Kitāb Munāqaḍāt man zaʿama annahu lā yanbaghī an yaqtaḍiya l-quḍāt fī maṭāʿimihim bi’l-a⁠ʾimma al-khulafāʾ, mentioning that this work is also attributed to al-Kisrawī al-Kātib, i.e., Mūsā ibn ʿĪsā.

3. Kitāb al-Aʿyād wa’l-nawārīz, the only work that would hint at an Iranian connection, although it probably contained Arabic poems on these feasts, lists of presents suitable at them in the Islamic period, etc.

4. Kitāb Murāsalāt al-ikhwān wa-muḥāwarāt al-khillān

In Yāqūt’s Muʿjam al-buldān, the only relevant⁹⁷ passage comes in the article on Tigris (ii: 440–442) (also mentioned in the article on Sātīdamā, iii: 169), where there is a lengthy (and seemingly freely paraphrased) quotation (via al-Marzubānī) from ʿAlī ibn Mahdī al-Kisrawī on the origin and course of Tigris, introduced by: “Abū ʿAbdallāh Muḥammad ibn ʿImrān ibn Mūsā al-Marzubānī: Abū l-Ḥasan ʿAlī ibn Hārūn gave me a sheet (waraqa) which he mentioned to be in the handwriting of ʿAlī ibn Mahdī al-Kisrawī.” The passage contains geographical information, but nothing specifically Iranian.

This summarizes the main relevant information on ʿAlī ibn Mahdī, who is much better known in the sources than his namesake.

As the bibliographical material shows, the works of these two al-Kisrawīs have been confused early on. At first sight, one would be tempted to attribute all the quotations related to Persian history to Mūsā ibn ʿĪsā, but the profusion of material on the nawrūz and the novelistic tendencies in the story of Balāsh may tip the balance in favour of ʿAlī ibn Mahdī, after all.

3.4 Ibn al-Muqaffaʿ and Nihāyat al-arab

Abū ʿAmr ʿAbdallāh (Rūzbih) Ibn al-Muqaffaʿ is the central character in the early process of translation from Middle Persian into Arabic and is one of the creators of Arabic literary prose.⁹⁸ Possibly⁹⁹ a convert from Zoroastrianism to Islam, he started his career in Nīshāpūr, serving the Governor Masīḥ ibn al-Ḥawārī (from 126/744), and continued it in Kirmān in the service Dāʾūd ibn Yazīd ibn Hubayra (130–131/748–749). Later, he lived mainly in Basra and Kufa.

Ibn al-Muqaffaʿ worked for the Umayyads and survived, for a short time, the takeover of the ʿAbbāsids before he was murdered in ca. 139/756.¹⁰⁰ Traditionally, he is said to have been no more than thirty-six at the time, but van Ess (1991–97) ii: 25, with good reason, sheds doubt on this. In addition to translations, he produced several works of his own, partly based on Persian materials, the best known among which is his Risāla fī l-ṣaḥāba.

Ibn al-Muqaffaʿ knew Middle Persian and had access to a variety of texts in that language. Some of his works are, in modified forms, extant, including the famous Kalīla wa-Dimna,¹⁰¹ although the transmission history of the text is extremely complicated and none of the preserved manuscripts can be taken as more than remotely reflecting the original.

The variety of Ibn al-Muqaffaʿ’s translations from, or works inspired by, Middle Persian materials, is impressive. Ibn al-Nadīm lists several of them in the passage dedicated to him in the Fihrist, p. 132/118//259–260:

1. Kitāb Khudāynāme fī l-siyar;

2. Kitāb Āyīn-nāme fī l-āyīn (Chapter 2.2.1);

3. Kitāb Kalīla wa-Dimna (Chapter 2.2.2);

4. Kitāb Mazdak (read: Marwak, see Chapter 2.2.1);

5. Kitāb al-Tāj fī sīrat Anūshirwān;

6. Kitāb al-Ādāb al-kabīr, known as *Mihr-jushnas(b) (MʾQRʾJSNS);¹⁰²

7. Kitāb al-Adab al-ṣaghīr;

8. Kitāb al-Yatīma fī l-rasāʾil, cf. also p. 364, sub Asmār al-Furs: Kitāb Rūzbih al-Yatīm;

9. Kitāb rasāʾilihi;

10. Kitāb Jawāmiʿ Kalīla wa-Dimna;

11. Kitāb risālatihi fī l-ṣaḥāba.

Of these, there are several that are of interest for the present theme, the translation of the Khwadāynāmag itself obviously leading the list. Several of the other listed works deal with wisdom literature, andarz, and are largely built on Middle Persian materials, though they are not translations of any particular work.

Al-Masʿūdī gives some titles of Ibn al-Muqaffaʿ’s other translations related to Persian national history, especially Kitāb al-Baykār and Kitāb al-Sakīsarān (cf. Chapter 2.2.1). Also the Arabic translation of the famous Nāme-ye Tansar, preserved only in a Persian retranslation, is attributed to Ibn al-Muqaffaʿ (Chapter 3.4). His activity as a translator and transmitter of Persian historical lore thus extended beyond the Siyar. This means that not even all the historical material explicitly circulating under his name needs to come from the Siyar and, through it, the Khwadāynāmag.

In some cases, the attributions are doubtlessly erroneous. Thus, e.g., Ibn al-Muqaffaʿ’s translations of Aristotle from Middle Persian seem to be doubly legendary: the translator in question was Muḥammad ibn ʿAbdallāh (ibn) al-Muqaffaʿ, i.e., the son of our Ibn al-Muqaffaʿ and, moreover, there is no indication that the material with which the latter worked would have been in Middle Persian.¹⁰³

The main point of interest for us is Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag. This translation is always mentioned first on the lists of Khwadāynāmag translations (cf. Chapter 3.1) and it is well documented in biographical and bibliographical sources.

In trying to grasp the contents of Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag, a major problem arises from the way sources quote his and other Arabic versions of the Khwadāynāmag. Ibn al-Muqaffaʿ’s translation is usually said to have been titled Kitāb Siyar al-mulūk, or Siyar mulūk al-ʿajam, and we may accept this as the original title.¹⁰⁴ It is rare to find direct quotations in extant sources, and even rarer that the microunits are explicitly quoted as coming from Ibn al-Muqaffaʿ’s Siyar. Most of the potential quotations (i.e., pieces of information that might derive from this book) are given with no indication of source: thus, al-Ṭabarī never explicitly quotes Ibn al-Muqaffaʿ in his Ta⁠ʾrīkh. His connection to Ibn al-Muqaffaʿ’s translation is entirely speculative. Sometimes the quotations are given without the translator’s/author’s name as coming from Kitāb al-Siyar, or Siyar al-mulūk – a title also borne by various other works and not necessarily referring to Ibn al-Muqaffaʿ – or, on the contrary, only quoted by the author’s name (and Ibn al-Muqaffaʿ wrote several works that might come into question). In very rare cases only, this is done with full indication of both the author and the title.

Under such circumstances, it is not easy to analyse the contents of the lost work. What does become clear from the unfortunately few explicit quotations from Ibn al-Muqaffaʿ’s Siyar is that, like most Arabic historical texts, it synchronized the Persian material with the sacred history of Islam.¹⁰⁵ Such elements certainly did not belong to the original Khwadāynāmag, as the Sasanians had no interest in discussing whether, e.g., Ḍaḥḥāk lived at the time of Noah or not. The synchronization must have been done by authors writing in Arabic for a Muslim readership.¹⁰⁶

There is one book that has been claimed to represent Ibn al-Muqaffaʿ’s Kitāb Siyar al-mulūk, at least to a certain extent.¹⁰⁷ This is Kitāb Nihāyat al-arab, an anonymous historical work, full of legends and concentrating on South Arabian history on the one hand, and on Persian national history on the other.¹⁰⁸ In the opening scene (Nihāya, p. 1), the Caliph Hārūn al-Rashīd, speaking with al-Aṣmaʿī, orders a Siyar al-mulūk to be brought forth from the Bayt al-ḥikma and al-Aṣmaʿī reads six parts (ajzāʾ) of it to the Caliph that very night. The book began with Sām ibn Nūḥ.¹⁰⁹

The Caliph asks al-Aṣmaʿī to collaborate with Abū l-Bakhtarī to produce a complete history of the world from Adam onward. The next morning the two scholars bring forth a Kitāb al-Mubtada⁠ʾ and proceed to compile from it and the Siyar a more complete work, the Nihāya itself. The Adamic prelude continues in the edition until p. 16, where the Siyar begins. This Siyar, it should be emphasized, is not primarily concerned with Persian history, which only comes into focus when the story has proceeded to Alexander and, especially, to the Sasanids. Its beginning is more concerned with South Arabian history (almost completely legendary and with little historical matter, except for the names of the rulers). Throughout the book, the two are synchronized, with the South Arabs at first as the focal point, only later conceding precedence to the Persians. Towards the end, the Islamic prehistory ousts the South Arabians from the focus and Mecca and the Quraysh take their place.

The preface of the Siyar (Nihāya, p. 17) tells us how two scholars, ʿĀmir al-Shaʿbī (d. 103/721) and Ayyūb ibn al-Qirriyya (d. 84/703), aided by a third, Ibn al-Muqaffaʿ (d. ca. 139/756), compiled the work by the order of the Caliph ʿAbd al-Malik ibn Marwān in 85/704.

This double preface, with its blatant anachronisms,¹¹⁰ does not lend credibility to the work, which could easily be passed by, were it not that in many cases it represents a text older than those of al-Ṭabarī and al-Dīnawarī, and the two can be shown to abbreviate the text of the Nihāya or its source.¹¹¹ This means that its core has to go back to a ninth- or perhaps even eighth-century original, even though the extant version clearly has undergone major modifications later.

The mention of Ibn al-Muqaffaʿ in the legendary preface is far from compelling evidence to accept the attribution.¹¹² Even if we did so, there would still remain the question as to whether the Ibn al-Muqaffaʿ quotations come from his translation of the Khwadāynāmag or some other book of his. In order to assess this, we have to take a close look at the material either explicitly or implicitly attributed to Ibn al-Muqaffaʿ in the book.

The first striking feature is that the Persian material, with few exceptions, only begins with Alexander the Great and strongly centres on the Sasanids. The legendary past of the Persian nation from the Creation onwards is lacking, except for a few minor notes and some synchronizations. Before Alexander, we only have brief mentions of Ḍaḥḥāk,¹¹³ Rustam (cf. Chapter 5.1) and Bahman/Dārā.¹¹⁴

A second striking feature concerns the contents of the stories purportedly taken from Ibn al-Muqaffaʿ. We find among them mainly things we would not expect to find in the Khwadāynāmag. To begin with, Alexander the Great is extensively discussed in a positive light. In later Persian (or Arabic) literature, this is not surprising as he became, thanks to the Alexander Romance, a legendary character and his mention in Surah 18 as Dhū l-Qarnayn cemented his fame.¹¹⁵ There are also strong Islamic features in this story (e.g., Nihāya, p. 128, Alexander’s pilgrimage to Mecca), but these, of course, could well have been added by Ibn al-Muqaffaʿ, who was catering to an Islamic audience.

In the native, Middle Persian tradition, however, Alexander is an accursed figure “that cursed Alexander the Roman” (ān gizistag Aleksandar ī Hrōmāyīg, see Chapter 2.3).¹¹⁶ The story in Firdawsī’s Shāhnāme is based on Islamic sources and should not be used as evidence for pre-Islamic Persian attitudes half a millennium earlier. As the Alexander story is explicitly quoted on the authority of Ibn al-Muqaffaʿ (Nihāya, p. 110), we either have to assume that the attribution is – and following this, the attributions in general are – purely fictitious or that Ibn al-Muqaffaʿ translated the Alexander story separately (of this we have no information whatsoever) or, finally, that if it really came from Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag, the author must have added extraneous materials to his translation: not only details, but substantial passages as well and these need not always come from Middle Persian sources.¹¹⁷

Thirdly, there are several long and elaborate stories among the material attributed to Ibn al-Muqaffaʿ and these, in fact, form the bulk of the Persian material in the book. Although we cannot completely exclude the possibility that the Khwadāynāmag did contain some long narratives,¹¹⁸ it seems that we should consider the Khwadāynāmag as a rather brief and dry work (see Chapter 6.1).

That the longer stories do not derive from the Khwadāynāmag is further supported by the fact that remarkably many of the more extensive stories in the Nihāya are known to have circulated as independent books. The following passages contain extended narratives related to Persian history:

1. pp. 82–85, Rustam, Isfandiyār, and Bahman: cf. Kitāb Rustam wa-Isfandiyār, translated by Jabala ibn Sālim (Chapter 2.2.1). Note that this quotation is introduced by the words “ʿAbdallāh ibn al-Muqaffaʿ has said: I found in the books of the Persians the (story of the) war between Rustam and Isfandiyār.” In Nihāya, p. 85, Ibn al-Muqaffaʿ is quoted as saying: “I found in Siyar mulūk al-ʿajam in the story of Bahman ibn Isfandiyār (wa-aṣabtu fī Siyar mulūk al-ʿajam fī qiṣṣat Bahman ibn Isfandiyār).” This would seem to locate the story of Isfandiyār within an Arabic Siyar. The story continues by telling how Bahman married Ūmīdh-dukht, the great-granddaughter of Solomon, together with a story about the rebuilding of Jerusalem, again material hardly deriving from the Khwadāynāmag.

2. pp. 110–158, Alexander: cf. Chapter 2.3.

3. pp. 161–171, Būdāsf: cf. Chapter 2.2.2 and Lang (1986).

4. pp. 177–200, Ardashīr and his ʿahd (pp. 197–200):¹¹⁹ cf. Chapter 2.2.1.

5. pp. 253–266, Bahrām Gūr: cf. Chapter 2.2.1.

6. pp. 277, 280–294, Balāsh and the daughter of the King of India: cf. Chapter 3.3.

7. pp. 294–346,¹²⁰ Qubād, continued by Kisrā Anūshīrwān, including the episodes of Mazdak, Anūshzād, and Buzurjmihr: cf. Chapter 2.2.1.

8. pp. 350–473, Bahrām Chūbīn, Kisrā Abarwīz, and the end of the Sasanid Empire (with intervening materials): cf. Chapter 2.2.1.

Excluding these, the Persian material is scanty and dull. This tallies well with our idea of the Khwadāynāmag as a rather concise chronicle. Most of the Sasanian biographies, excluding the ones above, are built of only three or four elements. To take a typical example, the short biography of Bahrām ibn Sābūr ibn Sābūr Dhī l-Aktāf (Nihāya, pp. 247–248) consists of four elements:

1. words spoken by him on ascending the throne;

2. a throne speech;

3. the sending of an encyclica (this element is missing from many short biographies);

4. a short report of his death and the number of his regnal years. In some biographical notes the towns founded by the king are added.¹²¹

Such concise entries perhaps best represent what the Khwadāynāmag might have looked like and they are fully in line with what Ḥamza, our best authority on the Khwadāynāmag, writes, as well as with the biographies of Agathias (Chapter 1.3.1). If that is the case, the novelistic materials would have to stem from sources other than the Khwadāynāmag and either Ibn al-Muqaffaʿ or some later author would have added them to the Siyar if the Siyar really is one of the sources of the Nihāya.

Here we have to consider the whole Arabic material. Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag was very influential, yet, e.g., Balāsh is virtually unknown in most sources that are supposed to have received material from the Khwadāynāmag translations (cf. Chapter 3.3). Likewise, Rustam, mentioned though not elaborated upon, in the Nihāya, is little known in Arabic books before al-Thaʿālibī (cf. Chapter 5.1). Had their stories been incorporated into the translation(s) of the Khwadāynāmag, they might be expected to have left more traces in the Arabic historical literature believed to have tapped the Book of Kings tradition. If, on the other hand, their stories only circulated in separate works less influential than Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag, their absence from historical works becomes unproblematic.

In the Nihāya itself, Ibn al-Muqaffaʿ once uses the formula “I found in the books of the Persians” (wajadtu fī kutub al-ʿajam, p. 82) and once “I read in the books concerning the lives of the Persian kings” (qara’tu fī kutub siyar al-mulūk min al-ʿajam, p. 159), instead of identifying any one specific book.¹²² Hence, one might argue that if these passages really come from him, they show the heterogeneous origins of his book. In both cases, the text continues with a long narrative (p. 82, Rustam; p. 159, the waṣiyya of Ādharwān, directly leading to the story of Būdāsf).

In addition to p. 85, cf. above, Ibn al-Muqaffaʿ twice uses the singular, namely Nihāya, p. 216 (wajadtu fī Kitāb Siyar al-mulūk) and p. 324 (innī wajadtu fī Kitāb Siyar mulūk al-ʿajam). In both cases this is followed by a short and concise passage. On pp. 216–217, there is a brief (six lines) biography of Narsī ibn Bahrām, and on p. 324, one sentence follows (“when he had ruled for thirty years, Kisrā Anūshīrwān took his troops and armies to Syria and conquered it”), which could well derive from the Khwadāynāmag. After it, there follows a long narrative concerning the cause of the war, in which Jabala ibn Ayham al-Ghassānī and al-Nuʿmān ibn al-Mundhir are involved. Again one doubts whether petty Arab kings were so important or interesting that they would have deserved a prominent place in the Khwadāynāmag. If we are to take this at face value, the war between Persia and Byzantium was caused by some camels having been abducted by one Arab tribal leader from another, which, clearly, is an Arab point of view. It is as if an Arabic author had fleshed out the dry framework of the original with related Arab lore which would be interesting only to his Arab patrons, not the rulers of the Sasanian Empire. The evidence is too meagre to be conclusive (and kitāb and kutub are easily confused in orthography), yet it may indicate a difference between the use of the singular, referring to the Khwadāynāmag, whether in translation or in the original, and the plural, referring to various Middle Persian sources, whether in translation or in the original, or it might even refer to Arabic compositions on Persian history. In any case, this shows that the Siyar is merely one among the putative Ibn al-Muqaffaʿ’s (numerous) sources.

There are indications that many of the long stories do not come from the Khwadāynāmag. The story of Bahrām Gūr is narrated on the authority of Ibn al-Muqaffaʿ (p. 256), yet it exhibits a strikingly Arab point of view – it could as well be called the story of al-Nuʿmān – which, again, one hardly expects to find in a Sasanian royal chronicle.¹²³ Slightly exaggerating, one could say that all longer narratives (the great, almost saintly Alexander; Bahrām Gūr and his Arab allies; the rebel heroes Bahrām Chūbīn and Anūshzād) transmitted on the authority of Ibn al-Muqaffaʿ in the Nihāya would have embarrassed the Sasanids and are, hence, out of place in their royal chronicle.¹²⁴

Fourthly and finally, there are some conspicuous similarities between the text of the Nihāya and certain passages, especially the story of Balāsh, that are elsewhere ascribed to al-Kisrawī, who may be the author of one Arabic version of Sasanian history, but could also be his namesake (Chapter 3.3). While it is quite possible that al-Kisrawī worked on the basis of Ibn al-Muqaffaʿ’s translation – there is no clear evidence either for or against such an assumption – it is significant that later sources quote al-Kisrawī and not Ibn al-Muqaffaʿ. As Ibn al-Muqaffaʿ’s Siyar must have been circulating more widely than the version of the obscure al-Kisrawī, we may conclude that it is not probable that an extended version of the story of Balāsh was already found in Ibn al-Muqaffaʿ’s work, from which it might be expected to have been quoted more widely than it actually is. If the al-Kisrawī referred to here was not the author of one of the Arabic versions of Persian national history, then there is no reason to ascribe these tales to the Khwadāynāmag in the first place.

This shows that parts at least of the Persian material in the Nihāya derive from a source later than Ibn al-Muqaffaʿ, so that even in the best of cases Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag is only one source of Persian national history for the author of the Nihāya, and the latter cannot be used for reconstructing either Ibn al-Muqaffaʿ’s Siyar or the Khwadāynāmag without first analysing its constituent parts.

The anonymous author of the Mujmal (p. 58/72) mentions having read the story of Balāsh and the daughter of the King of India from one Siyar al-mulūk, though he only summarizes the story in a few words. It is only found in the work of al-Kisrawī (quoted in ps.-al-Jāḥiẓ, Maḥāsin, pp. 242–251) and in the Nihāya, pp. 277, 280–294. Here we can be rather sure that the source for the Mujmal was either one of the two works or a common source of theirs. On the other hand, the Mujmal elsewhere explicitly identifies Ibn al-Muqaffaʿ as the author of the Siyar it uses (p. 2/2: Siyar al-mulūk az guftār o-rivāyat-e Ibn al-Muqaffaʿ). Two possible explanations arise: either the author is using a work which belongs to the tradition of the Nihāya (e.g., the Nihāya itself), where the attribution to Ibn al-Muqaffaʿ had already been made, or he is using a version of the Siyar elaborated by someone, e.g., al-Kisrawī, citing Ibn al-Muqaffaʿ as his authority. In the Nihāya, the story is narrated on Ibn al-Muqaffaʿ’s authority (p. 277), but it is extremely common to quote the ultimate, instead of the immediate, source.

All in all, it seems that the Nihāya, or its source, bases its narrative of Persian national history on a number of independent Arabic works, including novelistic stories of several semi-legendary heroes, known to have existed in Arabic as separate books. Whether or not these were translated by him, in the Nihāya they have summarily been attributed to Ibn al-Muqaffaʿ, as, e.g., in the case of Sīrat Isfandiyār (or Kitāb Rustam wa-Isfandiyār), which other sources attribute to Jabala ibn Sālim.¹²⁵ In addition, it probably uses the Arabic translation of the Khwadāynāmag by Ibn al-Muqaffaʿ or someone else. The Sasanian historical material is accurate enough to exclude any possibility of free fiction, which may well be behind much of the South Arabian material in the same book.¹²⁶

In assessing the position of the Nihāya in the Book of Kings tradition, we have to address two different questions, namely 1) does it represent, on any level, the Siyar by Ibn al-Muqaffaʿ?; and 2) does it represent, on any level, the Khwadāynāmag? The latter is easy to answer: the long stories probably do not derive from the Khwadāynāmag, whereas the concise royal biographies may well do so.

The first is a more difficult question. The Nihāya’s version of its own origin is legendary and anachronistic. On the other hand, it does have unique and accurate material which must go back to one or several reliable sources. Some of these may well be Ibn al-Muqaffaʿ’s “translations”, i.e., texts partly based on Middle Persian originals.¹²⁷ One of these sources, further, may well have been Ibn al-Muqaffaʿ’s Siyar, either as such or, perhaps more probably, in a version developed by some later author, such as al-Kisrawī.

3.5 Sources and Nature of These Translations

The large number of purported Arabic translations or versions makes it difficult for us to claim that the Khwadāynāmag was a large book, anything of the size of, e.g., al-Thaʿālibī, Ghurar. It would be a unique case in Arabic translation history that so large a book would have been translated several times between the eighth and the tenth centuries. The large number of translations makes it probable that the original was a rather brief text (see Chapter 6.2).

As far as we can see, there is no reason to assume that the translations were literal. That would go against the normal strategy of translating historical texts (Chapter 2.4) and there are clear traces of synchronization with the sacred history of Islam in the translations, which can hardly have been there in their Pahlavi original(s).

Baron Rozen (1895)¹²⁸ saw the various words used for versions or translations on Ḥamza’s list as technical terms. On this basis he divided the translations into three different groups, namely:

1. independent translations [naql] by Ibn al-Muqaffaʿ, Muḥammad ibn al-Jahm, and Zādūye ibn Shāhūye;

2. translations/compilations [naql aw jamʿ] by Muḥammad ibn Miṭyār and Hishām ibn Qāsim; and

3. redactions [iṣlāḥ] by al-Kisrawī¹²⁹ and Bahrām ibn Mardānshāh.

Rozen, and following him, all later scholars assumed these eight texts to have been translations of the Khwadāynāmag, which is highly improbable, see Chapters 3.1 and 3.2.1–6.

Rozen’s attempt to read a detailed difference between naql, jamʿ, and iṣlāḥ is, however, entirely hypothetical. Rubin (2008b): 44–45, and following him, Jackson Bonner (2011): 23, and n. 24, adopt Rozen’s theory assuming that the terms express clear differences and that they can be read as exact terminology distinguishing between three groups, the first term referring to translation proper, the second to compilation, and the third to “an edited reworking of material from various sources”. Rubin (2008b): 56–57, draws attention to the fact that Ḥamza himself uses the word (Ta⁠ʾrīkh, p. 22, quoting Bahrām al-Mōbadhānī) aṣlaḥtu in the sense “I established”. I would suggest that at least there the term means something like “I established a correct version either on the basis of several sources or by correcting the errors on the basis of knowledge derived from some source.”¹³⁰ Al-Kisrawī’s book (Chapter 3.3), on the other hand, seems either to have been a radical reworking of the original or, perhaps more probably, a completely new text, merely using the Khwadāynāmag as one of its sources. It seems hard to accept Rozen’s claim that the terms have been used in any exact and unvariable sense.

The translation history of both philosophical and scientific texts,¹³¹ on the one hand, and the Bible,¹³² on the other, shows many cases of translations which have later been edited by another scholar with or without comparison with the original. The same may be expected to have been the case of the Khwadāynāmag and there is no reason to assume that all authors on, e.g., Ḥamza’s list necessarily used any Pahlavi originals, though some may have done so.

3.6 Pre-Islamic Iran in Early Arabic and Persian Historical Texts

Very early on, pre-Islamic Iran found a firm place in the Arab world view.¹³³ Whereas Greece was more or less seen as a country of timeless philosophers and its history was neglected,¹³⁴ Iran and its history became an essential part of universal history for the Arabs. All historians writing in Arabic or Persian on general history included pre-Islamic Iran prominently in their books.

A problem of modern Arab-Islamic historiography is that the Iranian tradition is almost completely ignored. Thus, the legendary Kaʿb al-Aḥbār, who is credited with transmitting Jewish traditions into Arabic, receives an article in the gas (i: 304–305), while Ibn al-Muqaffaʿ is ignored, despite his translation of the Khwadāynāmag and other historical books that are better attested than the vague contributions of Kaʿb. The difference seems to arise from the material they were working with. Persian-based historiography is, perhaps, considered merely as translations and ignored, while the Biblical history – based on translations, too – is felt to be part and parcel of Arab-Islamic historiography.

Likewise, Schoeler (2002 and 2006) almost ignores Ibn al-Muqaffaʿ and the Persian tradition. In fact, however, the Persian-based Arabic world history seems to have developed early on, with the synchronization of Persian and sacred history. Early world historians either used the framework of Persian history, telling the story in the traditional Persian way, structuring their works on the Persian king lists, with the newcomers – the prophets and the sacred history – brought in through synchronization, or they adopted the Biblical model, probably inspired more than anything by Christian historical works. In these latter histories, the Persian kings were sometimes allotted a separate chapter, often ignoring the discrepancies ensuing from telling the world history twice, based on two different traditions, or the two traditions were synchronized but under the headings of the sacred history. Roughly speaking, one either puts the prophets under the respective headings derived from Persian national history (X was the prophet at the time of King Y) or the other way round (Y was the King at the time of the prophet X).

The Arab tradition of world histories was primarily based on Persian historiography and sacred history. In addition, a somewhat legendary South Arabian history was added to the repertory in some works, such as the anonymous Nihāyat al-arab. Graeco-Latin historical literature was largely ignored, except for what little trickled down through Christian Arabic historians.

The Persian material received by the Arabs mainly concerned Persian national history. There is no reason to assume that any Middle Persian historical text was interested in world history in the way we understand the term: as far as we know, Middle Persian historians only discussed the history of Iran, with its main adversaries, demons, Tūrānians, and Byzantines, merely finding their way into historical books for their battles against the Iranians. The element of world history was added by the Arabs, who synchronized the Persian material with other historical traditions available to them, namely the sacred history of Islam and the native Arab tradition, the Sīra of the Arab Prophet Muḥammad falling in between the two categories.

The mixture of these trends remained standard in later Arabic historiography. In Muslim Persian sources, obviously, the organization according to Persian kings remained more common than in Arabic sources. Even later, when some knowledge of Greek and Roman history gained entry into world histories, it remained in a marginal role and was mainly discussed as an extension of Persian history. Thus, e.g., passages on the Greeks in Orosius’ history were inserted into Persian history by the late Mamlūk historian al-Maqrīzī in his Kitāb al-Khabar ʿan al-bashar.¹³⁵

One of the first authors to write on world history outside of this frame was Rashīd al-Dīn Faḍlallāh (d. 1318), who added rather extensive¹³⁶ chapters on India, China, and Europe, listing both Popes and Emperors and synchronizing the two with each other in his Jawāmiʿ al-tawārīkh, thus creating a complete history of civilized nations as known to the Arab-Islamic culture. The various trends are, of course, rather imbalanced, as the last-mentioned three chapters, as well as the other “intruding” chapters, are rather uninformative lists of rulers, with occasional notes on them, whereas the traditional major trends (the sacred history, Arabic and Persian history) are fully developed.

The Arabs would, undoubtedly, have in any way created some framework for world history after they became acquainted with several historical narratives. However, Persian historiography was the first they encountered after having created their Empire and for this reason it was the Persians who had the greatest historiographical effect on them. Later, they would have had plenty of sources at hand for, e.g., Byzantine history, but they lacked the interest to take it fully into account. Only with the Nahḍa, the nineteenth-century renaissance, were other historical traditions fully absorbed into the Arab-Islamic worldview, but even then the basic structure of traditional Arabic historiography remained what it had been for more than a millennium, a combination of Arabic, Persian, and sacred history.

When it comes to pre-Islamic Persian history, it was the native Middle Persian tradition, the Khwadāynāmag among several other texts, that was the main source of information, which was supplemented by the few mentions of Persians in the Biblical history, mainly received from Christian sources, and the late Achaemenid history, which tied up with the Alexander Romance. Otherwise, there were few sources, such as Orosius, that gave the Arabs information on pre-Islamic Iran.¹³⁷

It is beyond the scope of the present work to compile a comprehensive list of all early authors whose works contain relevant materials, but this chapter introduces some of our main sources on pre-Islamic Persian history in Arabic and Classical Persian, works both lost and extant, in chronological order with some comments on each. Authors studied in more detail in Chapters 3.1–3.4 are not included here.

Hishām ibn Muḥammad al-Kalbī (d. ca. 206/821)

Hishām ibn Muḥammad al-Kalbī¹³⁸ is a major source for al-Ṭabarī, not only concerning pre-Islamic Iran but also more generally. He does not seem to have written much on pre-Islamic Iran, so that later sources which quote him on these matters probably received the information through oral channels.¹³⁹ Among the works attributed to him in Ibn al-Nadīm, Fihrist, pp. 109/96//208, one only finds a Kitāb Khabar al-Ḍaḥḥāk and a Kitāb Akhdh Kisrā rahn al-ʿArab, both of which – if they are genuine – were probably very short texts, taken down by his students, rather than fully developed monographs.¹⁴⁰ As a learned oral source for later historians Hishām is, though, invaluable.

Hishām’s sources, too, seem to have been mainly oral, rather than written. Thus in, e.g., Murūj §558, al-Masʿūdī refers to his sources by saying that he “narrated from his father and from other learned Arabs (…)”. It seems unwarranted to claim that any of his sources, or his sources’ sources, were necessarily written books: the early authors worked to a large extent orally and may well have received their knowledge from Persian Muslims (or non-Muslims) orally.

Abū ʿUbayda Maʿmar ibn al-Muthannā (d. 209/824)

Abū ʿUbayda had an interesting informant, ʿUmar Kisrā (Chapter 3.2.11). In Murūj §560, al-Masʿūdī mentions Abū ʿUbayda’s¹⁴¹ book on “akhbār al-Furs”¹⁴² – a term we might almost expect to describe a Siyar mulūk al-Furs. In this passage he describes the contents of the book:

Abū ʿUbayda Maʿmar ibn al-Muthannā has reported (dhakara) (information) from ʿUmar Kisrā in a book of his on the stories of the Persians (akhbār al-Furs) in which he describes the classes of their kings,¹⁴³ early and late, and the stories about them, their speeches, the divisions of their genealogies, the description of the cities they built and the districts they defined, the canals they dug and the noble families among them (ahl al-buyūtāt minhum) and how each group (farīq) of them marked themselves from among the Shahārija and others …

Al-Masʿūdī goes on to comment on the regnal years of the Petty Kings, which shows that ʿUmar Kisrā was also interested in chronology.

In §660, the relation between Abū ʿUbayda and ʿUmar Kisrā is made explicit:

Abū ʿUbayda Maʿmar ibn al-Muthannā has mentioned in his book on the stories of the Persians, a book he transmitted (rawāhu) from ʿUmar Kisrā …

The book which he transmitted from this ʿUmar is not preserved. Ibn al-Nadīm’s Fihrist does mention two books titled Akhbār al-Furs, one by Abū l-Ḥasan al-Nassāba Muḥammad ibn al-Qāsim al-Tamīmī¹⁴⁴ (p. 127/114//251: Kitāb Akhbār al-Furs wa-ansābihim), the other (p. 112/100//218) by al-Haytham ibn ʿAdī. But on the list of Abū ʿUbayda’s works (pp. 58–60/53–54//116–118), there is no book of this title. There is a Kitāb Khurāsān and another titled Kitāb Rawshanqubād (p. 60/54//117),¹⁴⁵ but neither of these would seem to be a general work on Persian history. There is, however, a third title, namely Kitāb Faḍāʾil al-Furs,¹⁴⁶ which will have to be considered.

In al-Qalqashandī’s Ṣubḥ al-aʿshā iv: 92,¹⁴⁷ there is a quotation from a book by this title, attributed to Abū ʿUbayd. This seems to be a mistake for Abū ʿUbayda, which is a common occurrence in Arabic texts. The contents of the quotation concern the building of Damascus by Bīwarasp and nicely fit the material transmitted by al-Masʿūdī. Even though the evidence is slight, it seems probable that the book in which Abū ʿUbayda transmitted material from ʿUmar Kisrā was his Kitāb Faḍāʾil al-Furs and this book should be considered a compilation of pieces of information on the early history of pre-Islamic Persia.

Muḥammad ibn Mūsā al-Khwārizmī (d. c. 232/847)

The mathematician and geographer Muḥammad ibn Mūsā al-Khwārizmī¹⁴⁸ is also credited with a Kitāb al-Ta⁠ʾrīkh, which is mentioned by Ibn al-Nadīm, Fihrist, p. 333/274//652.¹⁴⁹ Al-Masʿūdī, Murūj §8, mentions al-Khwārizmī among his sources, which probably refers to this book. Ḥamza, Ta⁠ʾrīkh, p. 145, quotes his own Kitāb Iṣfahān wherein he had quoted al-Khwārizmī’s Ta⁠ʾrīkh on earthquakes in 94 ah and 98 ah. The anonymous Tārīkh-e Sīstān, p. 95, quotes the book on the chronology of the birth of Prophet Muḥammad. This book probably concentrated on chronological material. How much information it contained on pre-Islamic Iran is not clear.

Abū Maʿshar al-Munajjim (d. 272/886 or later)

The astrologer Abū Maʿshar¹⁵⁰ was an authority on chronology and he used Persian, i.e., Pahlavi astrological works. He also dabbled with Hermetica¹⁵¹ and seems to have aimed at synchronizing various strands of history. His lost Kitāb al-Ulūf “Book of Thousands” probably included material on pre-Islamic Iranian chronology.

Ibn Qutayba (d. 276/889)

Abū Muḥammad ʿAbdallāh ibn Muslim al-Dīnawarī, better known as Ibn Qutayba, is one of the great scholars of the third/ninth century. Although his family was of Iranian origin, Ibn Qutayba was born in Iraq, but was later appointed the qāḍī of Dīnawar. From 257/871 until his death, Ibn Qutayba lived in Baghdad.¹⁵² Ibn Qutayba’s many works had a huge influence in various fields, but for the present purpose, two of them arise as the most important.

Kitāb ʿUyūn al-akhbār is a collection of anecdotes from various sources. Some of these concern pre-Islamic Iran, and Ibn Qutayba must have received some of the information orally, some from translations of Middle Persian texts – there is no indication that Ibn Qutayba himself would have been able to read Pahlavi. Ibn al-Muqaffaʿ seems to be the origin of much of this information, and he is at times cited summarily (qara⁠ʾtu fī kitāb li-Ibn al-Muqaffaʿ, ʿUyūn i: 54). Ibn Qutayba explicitly quotes from his Kalīla wa-Dimna,¹⁵³ al-Adab al-kabīr,¹⁵⁴ al-Tāj,¹⁵⁵ al-Yatīma,¹⁵⁶ Kitāb al-Āyīn,¹⁵⁷ and Kitāb Abarwīz ilā ibnihi Shīrūya.¹⁵⁸ He also quotes from Siyar al-mulūk, though the last is only clearly used four times as a book title¹⁵⁹ and could usually be translated as “lives of the kings,”¹⁶⁰ nor does it necessarily refer to Ibn al-Muqaffaʿ’s work, as Ibn Qutayba does not explicitly attribute it to Ibn al-Muqaffaʿ. In fact, there is nothing in the ʿUyūn that we would have reason to attribute to Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag. Usually, Ibn Qutayba cites his Persian sources very vaguely as “fī kitāb min kutub al-ʿajam”, or “fī kutub al-ʿajam”.¹⁶¹ Mostly he quotes sayings and other pieces of wisdom literature, with very little historical material. The information is usually not duplicated in his Maʿārif.¹⁶²

The other relevant work is his genealogical Maʿārif, which Ibn Qutayba ends with a chapter (pp. 652–667) on the Kings of Iran, giving as his source kutub¹⁶³ siyar al-ʿajam and adding a piece on the authority of Abū Ḥātim ← al-Aṣmaʿī (p. 652). The pre-Sasanian kings are mentioned only briefly and only a few kings are mentioned by name (Jam, Ṭahmūrath, Bīwarasf, Bahman ibn Isfandiyār, Dārā ibn Dārā, Alexander,¹⁶⁴ and the Petty Kings as a group). Several of them are synchronized with Biblical characters. Even the Sasanids are described only briefly. Several awkward events in the Sasanian history are discreetly passed by (Mani is not mentioned at all, Mazdak briefly on p. 663 as Mardaq; Bahrām Chūbīn, p. 664, is also mentioned only in passing, as Bahrām Shūbīna), although there are individual negative comments on some kings: e.g., Hurmiz ibn Narsī (p. 655) is said to have been gross and crude before his rule, and Yazdajird ibn Bahrām (pp. 659–660) is described in fully negative terms.

The biography of Sābūr ibn Ardashīr (Maʿārif, p. 654) may be taken as an example of Ibn Qutayba’s brevity, although it is far from being the shortest example:

Sābūr ibn Ardashīr. After him [Ardashīr] ruled his son Sābūr ibn Ardashīr, who adopted the ways of his father and his manners as to rigour and determination. He marched to Nisibis, where there were numerous troops of the Caesar. He besieged the city until he conquered it. After this he penetrated the Byzantine territory and conquered several towns before returning to his kingdom. He divided the prisoners-of-war between three towns, Gundīshāpūr, Sābūr in Fārs, and Tustar in al-Ahwāz. When he was about to die, he called his son Hurmiz and left the kingship to him, writing a contract (ʿahd) to him. He ruled in all 30 years and one month.

One of the main exceptions to the brevity of articles is Sābūr ibn Hurmiz Dhū l-Aktāf, discussed on pp. 656–659. The article contains the story of how before his birth his mother (not the mōbads, as in most versions) felt that the child would be a boy; Sābūr’s invention of one-way trafic as a child; how he got the title of Dhū l-Aktāf (Arab interest mentioned); his going disguised to Byzantium and getting caught and later escaping; and an extensive report of his building activities. Bahrām Gūr is also given a lengthy article (pp. 660–661), focusing on his deeds in India, told in an epic fashion,¹⁶⁵ while the Arabs are not even mentioned in the article.

The negative comments on some kings and the final narrative of the demise of the Sasanian kingdom (pp. 666–667) can hardly come from a Sasanian royal chronicle and Ibn Qutayba must have had other sources at his disposal (or these pieces had already been inserted by Ibn al-Muqaffaʿ). Otherwise, the general character of the chapter would fit the concise style of the Khwadāynāmag well.

Theodor Nöldeke (1879a): xxii, distinguished between two lines of transmission of Persian national history in Arabic and Classical Persian literature and took one of these to represent Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag, without actually having other basis for this than the references to Kitāb (easily confusable with kutub, also used by Ibn Qutayba in such contexts) Siyar al-ʿajam in Ibn Qutayba’s ʿUyūn al-akhbār.¹⁶⁶ While Ibn Qutayba does not explicitly attribute this work to Ibn al-Muqaffaʿ and while we know that a variety of Pahlavi historical books existed and an even larger variety of Arabic texts claimed to be, and sometimes were, translations of them, the identification is based on the simple misunderstanding that Ibn al-Muqaffaʿ’s translation is necessarily always the source for such information.

al-Yaʿqūbī (d. 284/897)¹⁶⁷

Al-Yaʿqūbī does not identify his sources for Persian history in the preserved part of his book – the lost beginning probably contained some information on them. The late position of Persian history in the book (Ta⁠ʾrīkh i: 158–177) seems significant: Persian kings are given only a minor role in his concept of world history, in contrast to most other world histories, where they either form the framework for the history (Nihāyat al-arab; al-Dīnawarī) or come second in importance, inserted into the framework of sacred history.

Al-Yaʿqūbī disparagingly glosses over the earliest Persian history (Ta⁠ʾrīkh i: 158: “The Persians claim for their kings many such things that cannot be accepted”, followed by some notes on Persian mythological and legendary figures, not identified by name) and then continues with a significant passage: “we have learned that they start counting the Kings of Persia from Ardashīr Bābakān onwards” (wa-wajadnāhum innamā yaḥsubūna mulk Fārs min ladun Ardashīr Bābakān). Before going on to these Sasanian kings, al-Yaʿqūbī prefaces them with a short list of 17 kings from Shayūmarth (sic) to Dārā with their regnal years and a mention (with no names) of the Petty Kings. All this is covered in less than a page.

Then al-Yaʿqūbī gives a rather dry chronological list of the Sasanian kings from Ardashīr to the last Yazdagird (i: 159–174), with some notes (i: 174–177) on their religion and geography. He only grows somewhat more verbous when he comes to Mani (i: 159–161), giving an unusual version of his career and death with a longish exposé of his doctrine and including the information that Sābūr first converted to Manichaeism. Bahrām Gūr’s story (i: 162–163) is closely linked to the Arabs. Khusraw Anūshirwān is also discussed somewhat more extensively (i: 164–165), and Bahrām Chūbīn is given a disproportionately long discussion (i: 166–172), almost as much as all the earlier kings put together. The last days of the Empire are briefly told (i: 172–174) and the chapter ends with various notes on Zoroastrianism, Iranian culture, Sasanian geography and administration, etc., clearly aimed at a non-Persian audience and, hence, not directly taken from any Pahlavi source, even though the information may ultimately come from there.

Grignaschi (1973): 125, has argued that al-Yaʿqūbī’s Ta⁠ʾrīkh constitutes a summary of Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag. This is speculative, as the book’s sources are not indicated, but it is by no means impossible, although one must keep in mind that the Khwadāynāmag, and hence presumably its Arabic translation, dealt with Persian history from the Creation onwards, while another Pahlavi book, translated into Arabic as Kitāb al-Ṣuwar, is known to have begun the story with the Sasanians. If this, or some other book on the Sasanians, was al-Yaʿqūbī’s main source, the scanty references to earlier kings might well derive from another source.

The only synchronization al-Yaʿqūbī gives, concerns Jamshād (sic; i: 20) and is given outside the chapter on the Persian kings. Likewise, some other pieces relevant for Persian history are given outside the chapter dedicated to it, and thus probably derive from other sources. Darius, Kisr Ḥūsh, and Artaxerxes are briefly mentioned as Babylonian kings (i: 82, 83). The table of contents of Kalīla wa-Dimna is given within Indian history, i: 88–89, and the story of chess and backgammon is related in the same chapter, i: 89–92, again in an unusual version, and Kitāb Makr al-nisāʾ (Sindbādnāme) is discussed in i: 93–94. Alexander is discussed both under Indian (i: 87–88, Porus) and Greek history (i: 143–145).

Although he does not indicate his sources in the preserved part of the Ta⁠ʾrīkh, in another work of his, the Buldān, p. 232 (preface), al-Yaʿqūbī mentions having collected oral historical material. The same may well hold true for his Ta⁠ʾrīkh, too.

al-Dīnawarī (d. not later than 290/902–3)

Abū Ḥanīfa Aḥmad ibn Dāʾūd al-Dīnawarī’s¹⁶⁸ al-Akhbār al-ṭiwāl¹⁶⁹ is a work which clearly differs in style from most other Arabic historical works of the time, except for the Nihāya, which it closely resembles in many parts. Its material sometimes shows similarities to that of al-Ṭabarī’s Ta⁠ʾrīkh, but its viewpoint is strongly Iranian, leading, e.g., to diminishing the role of the prophet Muḥammad, who is only mentioned in passing.

Al-Dīnawarī begins his book, p. 2, by saying that he will narrate the history of the kings of the world from Adam until the end of the rule of Yazdajird (and then listing the other nations and their rulers). As Adam is usually equated with Gayōmard, this closely follows the tradition of Persian history. Later, al-Dīnawarī synchronizes Persian, Biblical, and Arab histories to an extent few others have done, equating most of the central characters of the Persian tradition with those of the other two traditions.

Al-Dīnawarī mentions Ibn al-Muqaffaʿ on p. 9, but only as an ultimate authority whose information is transmitted orally (wa-yurwā anna Ibn al-Muqaffaʿ kāna yaqūlu), thus not directly referring to any source written by Ibn al-Muqaffaʿ. In general, al-Dīnawarī contains a good selection of Persian historical lore, which may partly go back to some translation of the Khwadāynāmag or some other historical work, but this material has undergone a profound modification, which makes it difficult to point to any specific sources.

Ibn al-Faqīh (wrote in 290/903 or soon after)

Abū Bakr Aḥmad ibn Muḥammad al-Hamadhānī, better known as Ibn al-Faqīh, wrote a geographical work titled Kitāb al-buldān, which was long considered lost, but an abbreviation of the book (Mukhtaṣar Kitāb al-Buldān) was printed early on.¹⁷⁰ It contains a great deal of material on pre-Islamic Persian history, but without reference to Persian books (in Arabic translation) which he may have used. At one point, p. 284, Ibn al-Faqīh quotes Ibn al-Muqaffaʿ on the genealogy of Ādhurbādh ibn Īrān ibn al-Aswad ibn Sām ibn Nūḥ (wa-yuqālu: Ādhurbādh ibn Bīwarasf – it is not clear whether this belongs to the quotation or not). This might come from Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag, which might imply that other materials on pre-Islamic Iran may also come from the same source. The genealogy derived from Noah, though, also shows that the latter part of the genealogy cannot come from Pahlavi sources.

Abū Muḥammad Dāʾūd Ibn al-Jarrāḥ (d. 291/903)

Dāʾūd Ibn al-Jarrāḥ,¹⁷¹ the grandfather of the Vizier ʿAlī ibn ʿĪsā, was of Persian extraction and al-Masʿūdī, Murūj §10, used his book, which contained “many stories about the Persians and other nations” (kitāb Dāʾūd ibn al-Jarrāḥ fī l-ta⁠ʾrīkh al-jāmiʿ li-kathīr min akhbār al-Furs wa-ghayrihim min al-umam).¹⁷² The title of this book may have been Kitāb al-umam al-sālifa.¹⁷³

Eutychius (Saʿīd ibn Baṭrīq) (d. 304/916)

In his Kitāb al-Ta⁠ʾrīkh,¹⁷⁴ which is organized according to Christian sacred history, Eutychius is well informed about Alexander (pp. 77–85) and the later Persian history, especially the Sasanians, which he interweaves with Christian history, but has little, if anything, to tell of earlier times that would derive from Persian sources. Mostly, the earlier Persians are mentioned through Biblical or Greek sources.¹⁷⁵ The only early Persian character that does not derive from Greek or Biblical historiography is Ṭaḥmūrat (p. 20), during whose times, Eutychius says, Zarathustra appeared, a detail which is rarely found in sources belonging to the Persian tradition, Zarathustra being usually dated to the reign of Gushtāsb (e.g., al-Ṭabarī, Ta⁠ʾrīkh i: 648//iv: 46). The confusion arises from the identification of Zarathustra with Būdāsf, who is often dated to Ṭahmūrath’s reign (e.g., Ḥamza, Ta⁠ʾrīkh, p. 27).

The Askānians, identified as the Petty Kings, are briefly mentioned on p. 85. It is only with the founder of the Sasanid dynasty, Ardashīr¹⁷⁶ (pp. 106–108), that Eutychius starts presenting Persian material ultimately derived from Persian sources, and this continues until Kisrā Abarwīz (pp. 213–218).¹⁷⁷ Some major events are discussed more extensively, but the majority of kings are passed by with rather short notes. As the author himself mentions that he has abbreviated his sources (wa-jaʿaltuhu mukhtaṣaran, p. 3), this does not necessarily mean that the sources he used were equally concise. The major exceptions to this brevity are the stories about Bahrām Gūr (pp. 176–179), Kisrā Anūshirwān (pp. 207–210), and Kisrā Abarwīz and Bahrām Chūbīn (pp. 213–218), all of which we know to have circulated as separate books (Chapter 2.2.1). At the end of the last story, Kisrā is said to have converted to Christianity, which is a strong indication that this story does not come, at least not directly, from Middle Persian sources. There is no mention in the book of the Sistanian heroes.

Eutychius does not tell us anything about his Persian sources. Nöldeke put forward the idea that Eutychius’ source would have been Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag and used Eutychius to reconstruct the contents of that book. This is possible, and there are similarities between Eutychius and what we know about the Khwadāynāmag. As shown in Chapter 6.2, the articles of the Khwadāynāmag on the various kings were most probably short, and the Sistanian heroes may not even have been mentioned there (Chapter 5.1), just like they are absent in Eutychius.

There is, however, one major difference. The evidence strongly points to the Khwadāynāmag as having begun from Gayōmard and gone through the mythic and legendary kings of Iran before Alexander the Great. Quotations from Ibn al-Muqaffaʿ, too, seem to indicate that they were present in his translation (Chapter 3.7) and that he synchronized this history with the sacred history. We cannot readily see why Eutychius would have opted not to quote these passages. He presents ancient Persian history through Greek historiography, mentioning the Achaemenids, but this cannot be the reason for omitting the legendary Persian kings, because they are usually synchronized with Biblical prophets with whom the Achaemenids had little to do, and thus could have found a niche of their own in Eutychius’ history. The emphasis on Sasanian history makes it possible that his source was another book that only discussed the Sasanians, such as Kitāb al-Ṣuwar (Chapter 2.2.1).

Be this as it may, Eutychius does derive his information on the Sasanids rather directly from (translated) Middle Persian sources, i.e., in a form that adds little Arabic material or material that would be problematic for Sasanian sources. Thus, Mani is mentioned as a mere rebel (p. 111), without the embarrassing stories about how the kings first favoured him;¹⁷⁸ the name of Sābūr Dhū l-Aktāf is explained without reference to the Arabs particularly (p. 115: “He was called Dhū l-Aktāf [the One of the Shoulders] because when he vanquished some king he dislocated his shoulder”).¹⁷⁹ Likewise, al-Nuʿmān is not mentioned in the main story of Bahrām Gūr (pp. 176–178), only in a brief (mere three lines) end note, where it is said that “some Persians¹⁸⁰ mention that Bahrām Jūr was in the care of al-Nuʿmān ibn al-Mundhir al-Lakhmī, the King of the Arabs in the desert (…).” That the Arabs are not mentioned is precisely what one would expect from a Middle Persian source: for the Sasanids, the Arabs were not the centre of interest.

Eutychius has much common material with al-Ṭabarī (e.g., pp. 190–191 on Qubād, cf. al-Ṭabarī, Ta⁠ʾrīkh, i: 883//v: 128ff.), which shows that they used a common source (or common sources), but as al-Ṭabarī is much better informed, this common source is probably just one of al-Ṭabarī’s sources.

Eutychius mentions that Khusraw Anūshirwān “took out” (probably meaning that he put them in circulation) the “books” of Ardashīr wherein there was his sīra “way of life” (wa-akhraja kutub Azdashīr allatī fīhā sīratuhu allatī sāra bihā), made people follow this sīra, and wrote about this to the four corners of the world. This seems to come from Eutychius’ source and there is no reason to assume that Eutychius himself would have been familiar with the book. This seems to refer to an andarz book, presumably containing wise sayings attributed to Ardashīr (cf. Chapter 2.2.1).

Ibn Khur(ra)dādhbih (d. c. 300/912)

Better known for his Kitāb al-Masālik wa’l-mamālik, the geographer Abū l-Qāsim ʿUbaydallāh ibn ʿAbdallāh (or Aḥmad) Ibn Khur(ra)dādhbih (d. c. 300/912), of Iranian origin and a convert from Zoroastrianism, is also credited with a Ta⁠ʾrīkh and a Kitāb Jamharat (Jumhūr) ansāb al-Furs wa’l-nawāqil.¹⁸¹ The Ta⁠ʾrīkh is quoted by al-Thaʿālibī, Ghurar, pp. 130–131, and Ibn Khurradādhbih, without mention of the title al-Thaʿālibī is referring to, is further mentioned as an authority or quoted on pp. 257, 263, 378, 415–416, 486, 556–557,¹⁸² and 604–605.¹⁸³ Kitāb Jamharat ansāb al-Furs wa’l-nawāqil was also used by al-Masʿūdī in his Murūj §503, who informs us that it was concerned with pre-Islamic nations (dhikr al-umam al-māḍiya qabla majī’ al-Islām).¹⁸⁴

al-Ṭabarī (d. 310/923)

Abū Jaʿfar Muḥammad ibn Jarīr al-Ṭabarī¹⁸⁵ is arguably the most important historian who wrote in Arabic. Born in 224 or 225/839, he died in Baghdad in 310/923. He lived his early life in Iran, before moving to Iraq when he was already in his teens. After this, he travelled in various Arab countries before settling in Baghdad. Al-Ṭabarī clearly understood Persian, but there is no indication that he could have read Pahlavi. A large Qur’ānic commentary, Tafsīr, is his main religious work and is hugely influential for all later tafsīr until the present day. His (Mukhtaṣar) Ta’rīkh al-rusul wa’l-mulūk is a gigantic world history, ranging from the Creation to the early fouth/tenth century.

The work begins with the Creation and the earliest prophets, but soon interweaves pre-Islamic Persian history into the grid arranged according to prophets – as the title of the book intimates, the prophets come first in order and kings are arranged according to them until the birth of Islam. From the institution of the Islamic (Hijrī) calendar onward, the book changes into an annalistic form.

While much of the Islamic history is transmitted in the Ta⁠ʾrīkh in the khabar form as short narratives and with authorities quoted for each passage, Persian history is amalgamated into a continuous narrative, which is then narrated in sections interspersed with other events elsewhere, which al-Ṭabarī considered contemporaneous (especially sacred history and South Arabian history). Al-Ṭabarī does not usually indicate his sources for this part, mainly referring to Ibn al-Kalbī as an authority when he does so. He is also known to have used the text preserved in ms-Sprenger. No Persian books or their translations are mentioned in the Ta⁠ʾrīkh.

As discussed in Chapter 1.1.2, Theodor Nöldeke suggested that the Khwadāynāmag was one of the major sources for this part of al-Ṭabarī’s book. While this theory cannot be substantiated, it is clear that al-Ṭabarī used several Pahlavi texts in Arabic translation either directly or through earlier Arabic compilations, though we cannot identify them with any certainty. Though there is no unequivocal evidence for it, the Khwadāynāmag may well have been one of them, but there is also reason to believe that other texts known to have existed in Arabic translation, such as Kitāb al-Ṣuwar, Kārnāmaj Ardashīr, some version of the story of Bahrām Chūbīn, and perhaps a translation of Ayādgār ī Zarērān (Chapters 2.2.1 and 4.6), were familiar to him. As al-Ṭabarī was fully able to use several sources in other parts of his Ta⁠ʾrīkh, it would be absurd to claim that all his material on pre-Islamic Iran would need to come from one single source.

al-Masʿūdī (d. 345/956)

Abū l-Ḥasan ʿAlī ibn al-Ḥusayn al-Masʿūdī wrote two books that are well informed in Persian matters and are still extant, Murūj al-dhahab and Tanbīh wa’l-ishrāf.¹⁸⁶

In the Murūj, the main section on Persian history comes in §§530–663, but elsewhere, especially in the first two volumes, there are many scattered pieces of relevant information. In §§ 8–14, al-Masʿūdī lists a total of 83 earlier authors or authorities he has used. Among them one finds Sahl ibn Hārūn, ʿAbdallāh ibn al-Muqaffaʿ, and Muḥammad ibn Mūsā al-Khwārizmī (§8), Ibn Khurradādhbih and al-Dīnawarī (§9), Dāʾūd ibn al-Jarrāḥ and Kitāb al-Ta⁠ʾrīkh al-jāmiʿ li-funūn al-akhbār wa’l-kawā⁠ʾin fī l-aʿṣār qabla l-islām wa-baʿdahu, written by Abū ʿAbdallāh Muḥammad ibn al-Ḥusayn ibn Sawwār, known as Ibn Ukht Abī ʿĪsā ibn Farrukhān-shāh, which continued until the year 320 (§10), as well as Ibn Qutayba’s Maʿārif and al-Ṭabarī (§11).

Al-Masʿūdī also knew ʿAlī ibn al-Jahm’s qaṣīda “fī badʾ al-khalq”, written in rajaz muzdawija of 12 syllables (§49),¹⁸⁷ as well as an anonymous (qāla dhū ʿināya bi-akhbār al-ʿālam wa-mulūkihi)¹⁸⁸ poem (in monorhyme basīṭ in -ānū) discussing, or perhaps only listing, the titles of the kings of the world, their kingdoms, and their names (jumalan min marātib mulūk al-ʿālam wa-mamālikihim wa-asmāʾihim), of which he quotes six verses. In §503 he further mentions Ibn Khurradādhbih’s geographical al-Masālik wa’l-mamālik and his book on Ta⁠ʾrīkh, as well as a world history attributed to Aḥmad ibn al-Ṭayyib ṣāḥib al-Muʿtaḍid [al-Sarakhsī].

In addition to a large selection of prose works, al-Masʿūdī had some historical poems at his disposal or was at least aware of them. As far as al-Masʿūdī is concerned, he seems to have only used Arabic sources, either original compositions or translations, and when he quotes verses from such poems they are in Arabic. Others, however, may have been in some form of Persian, either in a written form of Middle Persian, or if oral, in some form of very early Persian, a language form which is sparsely documented. Al-Masʿūdī, Murūj §538, quotes 4 ramal verses, rhyming in -aCam and mentioning how “we” divided the world between Sal(a)m, Ṭūḥ (or Ṭūj), and Īrān, the three sons of Afrīdūn.¹⁸⁹ The verses are attributed to a Persian poet who had lived in the Islamic period (baʿḍ al-shuʿarāʾ mimman salafa min abnāʾ al-Furs baʿda l-islām yadhkuru wuld Afrīdūn al-thalātha). The verses may well be a self boast (iftikhār) from a qaṣīda, but they could also be a fragment of an epic, a genre that is not completely lacking in Classical Arabic literature.¹⁹⁰ Likewise, there are in Murūj §608 seven basīṭ lines of narrative poetry, rhyming in -ārī and attributed to “an early Persian poet” (baʿḍ al-mutaqaddimīn min al-shuʿarāʾ min abnāʾ al-Fārs). It is noteworthy that these Persian poets wrote in Arabic. Hence, references to epic poetry should not without further study be taken as indicative of Persian poetry.

Although not listing them at the beginning of his book, al-Masʿūdī also had access to several translations of Middle Persian literature by Ibn al-Muqaffaʿ and others. For a discussion of these sources mentioned in Murūj §§479–480, 541, 543, and 644, as well as Tanbīh, p. 106//150–151, see Chapter 2.2.1. He is also our only source for quotations from ʿUmar Kisrā (Chapter 3.2.11).

In all, al-Masʿūdī is well aware of Persian history from the Creation onwards and presents a wide selection of the material that later found its way into Firdawsī’s Shāhnāme, and in some cases he seems to have discussed these more extensively in his lost books.¹⁹¹

Tanbīh, pp. 85–111//122–158, partly covers the same material with a somewhat heavier emphasis on chronology. It also contains references to a variety of Pahlavi books translated into Arabic, including the Khudāynāmāh, Āyīnnāmāh, and Kitāb al-Ṣuwar, the last only being described without a mention of the book’s original title (see Chapter 2.2.1).

Ḥamza al-Iṣfahānī (d. 350/961 or 360/971)

Ḥamza ibn al-Ḥasan (or al-Ḥusayn) Abū ʿAbdallāh al-Iṣfahānī was a learned philologist, living in Isfahan and known to have visited Baghdad.¹⁹² In his various works, he quotes, or refers to, several Arabic historians, including Muḥammad ibn Mūsā al-Khwārizmī, Ta⁠ʾrīkh; Ibn Qutayba, Maʿārif; and Abū Maʿshar al-Balkhī, Ikhtilāf al-ziyaja and Kitāb al-ulūf.¹⁹³ For the sources he quotes in his Ta⁠ʾrīkh sinī l-mulūk, see below.

Ḥamza wrote two historical works.¹⁹⁴ One of them, Ta⁠ʾrīkh Iṣbahān, also referred to as Kitāb (al)-Iṣfahān, has been lost except for fragments,¹⁹⁵ but the other, Ta⁠ʾrīkh sinī l-mulūk, has been preserved¹⁹⁶ and is a most important source for the study of the Khwadāynāmag, but there is no indication that the author himself would have known Pahlavi (cf. below).¹⁹⁷ In addition, his lost Risāla fī l-ashʿār al-sāʾira fī l-nayrūz wa’l-mihrajān is quoted by al-Bīrūnī, Āthār, p. 38/31//36.

An important quotation from Kitāb al-Iṣfahān is found in Mujmal, p. 40/47 (about Kursī-ye Sulaymān, claimed to have been built by demons on the order of Sulaymān at the request of Kay Kāwūs):

Ḥamza al-Iṣfahānī rejects this concerning the Kursī and in his Kitāb al-Iṣfahān explains that on these stones there are many pictures of hogs, which are more inimical (dushmantar) in the eyes of the Israelites than any other animals. In the site, there are inscriptions in Pahlavi, and he goes on to tell that once a mōbad was brought to read them. Among them was the following: “He built this house of Jam in such-and-such a day of so-and-so a month.”¹⁹⁸ This, as well as much more, is written in Pahlavi. I have not copied (this) because I do not know their letters, so that no rancour (against me) would arise from their form. They call this The Thousand Pillars (hazār-sutūn).

It seems probable that the comment on not knowing the Pahlavi script comes from Ḥamza, although theoretically it could also be by the anonymous author of the Mujmal. If, as it would seem, this comes from Ḥamza, it makes it abundantly clear that he did not know any Pahlavi.¹⁹⁹ The passage also shows the acumen of Ḥamza, rejecting a claim on the basis of personal inspection of the site.

Ta⁠ʾrīkh sinī l-mulūk, written in 350/961 or a year after,²⁰⁰ consists of ten chapters,²⁰¹ the first of which concerns pre-Islamic Iran. Ḥamza had access to several translations of the Khwadāynāmag and other historical books, which makes him an important witness for them (see Chapter 3.1).

The chapter on pre-Islamic Persia has the following structure:

1. a general introduction, seemingly based on a variety of sources, of which only Abū Maʿshar is quoted by name. He seems to be the main source for this subchapter. The List of Ḥamza is given in this part (pp. 9–15);

2. a long, mainly chronological quotation from Mūsā ibn ʿĪsā al-Kisrawī (pp. 16–21);

3. a long chronological quotation from Bahrām ibn Mardānshāh (pp. 22–25);

4. narratives about Persian kings from Ūshahanj to Yazdajird ibn Shahriyār. The sources for these stories (akhbār)²⁰² are given in general terms, such as (baʿḍ) kutub al-Siyar (pp. 26, 27, 30), baʿḍ al-nusakh (p. 26), kutub²⁰³ al-ʿarabiyya (p. 28), baʿḍ al-ruwāt (p. 30), wa-fī mā walladahu²⁰⁴ l-quṣṣāṣ min al-akhbār (p. 33), Kitāb Ṣuwar mulūk Banī Sāsān (several excerpts on pp. 38–49). Much of this material is taken from books other than kutub al-tawārīkh wa’l-siyar (p. 49, see translation in Chapter 7.3) (pp. 26–49);²⁰⁵

5. a story claimed to derive from the Avesta²⁰⁶ and another version of the same story without attribution to any source, the latter (or both) possibly from Abū Maʿshar (see Chapter 3.6) (pp. 50–51).

In the last chapter heading (p. 50), Ḥamza seems to imply that his main sources for the Khwadāynāmag were Ibn al-Muqaffaʿ and Ibn al-Jahm: “Chapter Five of the first Book narrating things which are in the Khudāynāme but which Ibn al-Muqaffaʿ and Ibn al-Jahm did not relate.”²⁰⁷ Then he gives the passage which he had “read in a book translated from a book of theirs entitled al-Ābistā (the Avesta).” It should be noted that he does not say anything about his other sources, of which only Mūsā ibn ʿĪsā and Bahrām ibn Mardānshāh are quoted in the book and were thus certainly used by Ḥamza. It may well be questioned whether Ḥamza had, in fact, had at his disposal all, or even any, of the remaining books he lists or whether he, too, is merely copying some older source or name-dropping titles that he knew.

Ḥamza also quotes from Abū Maʿshar al-Munajjim, Kitāb al-Ulūf, and refers to Ibn Qutayba’s Kitāb al-Maʿārif (pp. 77, 82).²⁰⁸ Hishām ibn al-Kalbī is also often mentioned (e.g., p. 83), although not in the chapters on Persia. The same goes for Muḥammad ibn Jarīr al-Ṭabarī (p. 117: fī kitābihi l-musammā al-Kitāb al-Mudhayyal).²⁰⁹

The note on Bahrām Gūr (Jūr) (p. 43) ignores any Arab aspects of the story and instead concentrates on telling how he introduced Indian music into Iran. The Persian section is narrated without synchronizations with the sacred history, although later in the book, Ḥamza quotes stories narrated by (wa-qara⁠ʾtu fī akhbār rawāhā) ʿĪsā ibn Dāb²¹⁰ synchronizing Jam(shīd) and Hūd (pp. 97–98),²¹¹ as well as other characters of Persian and sacred history, though being himself rather critical towards ʿĪsā. The only note on Rustam, synchronized with South Arabian history, comes from the chapter on South Arabian kings, and on pp. 103, 104, there are two further synchronizations taken from kitāb min kutub akhbār al-Yaman. The same synchronization continues in the chapter concerned with the birth of Islam, based on al-Ṭabarī’s al-Kitāb al-Mudhayyal.

This gives more credence to his Persian part being directly derived from Middle Persian sources in translation, lacking elements that are often present in Arab-Islamic historiography but are highly unlikely to derive from Pahlavi sources.

The anonymous Mujmal quotes Ḥamza’s Ta⁠ʾrīkh extensively, sometimes without indication of the source. In some cases, the text of Mujmal is superior to the edited text of the Ta⁠ʾrīkh, and a detailed comparison of the two might help us improve on Ḥamza’s edited text.

al-Maqdisī (d. after 355/966)

Muṭahhar ibn Ṭāhir al-Maqdisī’s²¹² (alive in 355/966) Kitāb al-Badʾ wa’l-Ta⁠ʾrīkh is a universal history which uses a wide selection of written sources,²¹³ preserves quotations from lost works, and includes unique information derived from oral sources. The author’s life is little known and our main source on him is the Badʾ itself. Al-Maqdisī was writing in Bust under the commission of a Sāmānid vizier.²¹⁴ The Badʾ probably dates from 355/966, the year which is occasionally referred to in the book as the present (Badʾ i: 6; ii: 152).²¹⁵

The Bad’ is his only preserved work, but in it al-Maqdisī refers to another work of his, a Kitāb Maʿānī l-Qur’ān (e.g., Badʾ ii: 95). Al-Maqdisī used Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag (Chapter 3.7),²¹⁶ Ibn Qutayba’s Maʿārif (Badʾ ii: 150), and Ibn Khurradādhbih’s Ta⁠ʾrīkh (Badʾ ii: 151, vi: 51, 89) and Masālik (Badʾ iv: 19, 61). He also refers to something that he had read “in some of the Persians’ siyar”: the information he gives primarily concerns the synchronization of Persian and sacred history and cannot, thus, come directly from a Pahlavi book.

Al-Maqdisī tells how he visited an ancient fire temple (bayt nār) in Khūz, a district of Fārs. He did this in order to ask the local Zoroastrians a question. Subsequently, some leaves (ṣuḥuf) of a book called the Avesta (al-Abisṭā) were brought forth to provide him with an answer (Badʾ i: 62–63). In Badʾ ii: 59–60, he may be referring to the same informant whom he here identifies as hirbadh al-Majūs, “the Zoroastrian hērbad”.²¹⁷ He also discussed some points with a man belonging to the Zoroastrian sect of Bihāfarīdiyya (Badʾ i: 176).²¹⁸

Al-Maqdisī is able to quote Persian at first hand. For the older forms of the language he most probably depends on Zoroastrian scholars, but for Classical Persian he does well on his own. He is able to quote a few verses from the Persian historical poem (qaṣīda) by al-Masʿūdī al-Marwazī (Badʾ iii: 138, 173 – Chapter 4.1.1), which not only shows that he was acquainted with the language but that he considered his readers, too, to be able to understand it. Al-Maqdisī is also able to explain the meaning of Persian words (e.g., Badʾ i: 63). Another indication of his familiarity with Persian is the story related in Badʾ iii: 188–195. The same story is also told by Ibn Hishām (Sīra i: 69–73),²¹⁹ Ibn Qutayba (ʿUyūn i: 236–237, abbreviated, and introduced by: qara⁠ʾtu fī kutubi l-ʿajam),²²⁰ and al-Ṭabarī (Ta⁠ʾrīkh i: 946–950//v: 236–242). All versions differ from each other in wording and details but agree in the general story line, if we ignore the radical abbreviation of the story by Ibn Qutayba, but only al-Maqdisī’s version contains Persian expressions missing from the others (e.g., Badʾ iii: 192 fa-qāla bi’l-fārsiyyati: īn kūdhak khar-ast, yaʿnī: ibnu l-ḥimār; cf. al-Ṭabarī, Ta⁠ʾrīkh i: 949//v: 240, qāla: ibnatu l-ḥimār; Ibn Hishām, Sīra i: 71 qāla Wahriz: bintu l-ḥimār; missing from Ibn Qutayba).

Miskawayh (d. 421/1030)

Abū ʿAlī Aḥmad ibn Muḥammad Miskawayh (d. 421/1030) was a philosopher and historian who served several Viziers in Iran.²²¹ His main historical work is the world history Tajārib al-umam, which continues until 369/980. His anthology of wisdom texts, al-Ḥikma al-khālida, contains a wealth of Persian materials.

The Tajārib begins, after a two-page Introduction (i: 59–60), with Persian history as the organizing principle until the history of the Prophet Muḥammad takes over in i: 169. Most of the intervening 110 pages are concerned with Persian history. ʿAhd Ardashīr is reproduced in its entirety (i: 97–107) and Sīrat Anūshirwān, allegedly written by Anūshirwān himself, is presented in large extracts (i: 132–142). The same interest in wisdom literature is seen throughout the section on Persian history. Unfortunately, Miskawayh does not usually indicate his sources.

al-Bīrūnī (d. about 442/1050)

The famous polymath Abū Rayḥān Muḥammad ibn Aḥmad al-Bīrūnī,²²² who served the Samanids and the Ghaznavids, is one of our best sources for Persian history and he had many otherwise unknown sources both in Arabic and Persian at his disposal. Al-Bīrūnī himself wrote exclusively in Arabic. A prolific writer, al-Bīrūnī’s main work for information on Persian national history is his history of ancient nations al-Āthār al-bāqiya ʿan al-qurūn al-khāliya, written about 390/1000. His book on India, Kitāb fī taḥqīq mā li’l-Hind, written about 420/1030, also occasionally provides information on Iran. The Āthār is particularly valuable, as al-Bīrūnī used several sources in early Classical Persian. These will be discussed in Chapter 4.1.

Gardīzī (wrote in early 440s/1050s)

Little is known about the life of Abū Saʿīd ʿAbd al-Ḥayy ibn Ḍaḥḥāk ibn Maḥmūd Gardīzī. He seems to have been in close contact with the Ghaznavid court and he dedicated his main work, Zayn al-akhbār, to the Sultan ʿAbd al-Rashīd ibn Maḥmūd, who ruled in 440–443/1049–1052.²²³ The work has only been partially preserved.

Among the books Gardīzī quotes are Ibn Khurradādhbih’s Akhbār (Zayn, p. 370, presumably the same as Ta⁠ʾrīkh) and a work titled Kitāb Rubʿ al-dunyā or Tawḍīḥ al-dunyā, attributed by him to Ibn al-Muqaffaʿ (Zayn, pp. 370, 402, on Turks). The part on Persian history (Zayn, pp. 65–105) is concise and lacks any indications of sources.

Tārīkh-e Sīstān (main part written soon after 448/1062?)

Tārīkh-e Sīstān is a modern conventional title, and the real title of this anonymous work may have been Faḍāyil-e Sīstān.²²⁴ The book consists of several layers, the main part going back to around 448/1062.²²⁵ This oldest layer uses only old sources and is valuable for source critical studies. The old sources²²⁶ related to Persian national history that the author either mentions or quotes include a Kitāb-e Faḍāyil-e Sīstān (p. 49) written by an unknown Hilāl-e Yūsuf-e Awqī(?). He also refers to an older book on Sistan, Akhbār-e Sīstān (p. 56), as well as to Ibn al-Muqaffaʿ’s Kitāb-e Siyar-e mulūk-e ʿajam (p. 56), both on the building of Arak in Sistan by Alexander the Great.

Further, Tārīkh-e Sīstān mentions that “Bū’l-Muʾayyad-e Balkhī and Bishr-e mqsm say in Kitāb-e ʿAjāyib-e barr o-baḥr” (p. 58),²²⁷ continuing with a quotation on the wonders of Sistan. Abū l-Muʾayyad (Chapter 4.1.3) is also given as the authority for a quote from the Bundahishn (Kitāb-e Ibn Dahshatī)²²⁸ on the other wonders of Sistan (pp. 60, 61) and he is quoted as presenting yet another wonder on his own authority (p. 61). The quote probably comes from the same book on wonders. On p. 75, there is a quotation from Abū l-Muʾayyad’s Kitāb Garshāsb,²²⁹ relating the story of Kay Khusraw’s travel to Azerbayjan (Ādharbādgān) with Rustam, where Adhurgush[n]asb came to bring a light to the darkness.

There is also a reference to a Shāhnāme, which tells the stories of Narīmān, Sām, and Dastān (p. 53). The anonymous author does not identify the author of this Shāhnāme, but it is not Firdawsī, as the passage continues: “the story of Rustam is among those which Bū l-Qāsim Firdawsī versified in the Shāhnāme”, clearly speaking of two different works. It is possible that the Shāhnāme here refers to Abū l-Muʾayyad’s Shāhnāme.

The author also mentions Akhbār-e Farāmarz in twelve volumes (p. 53), probably referring to the prose original of the later versified epics, see Chapter 4.7. From among the nāme literature, he also mentions a Bakhtiyārnāme, containing the story of Bakhtiyār “from among the children of Rustam” (p. 54), again probably the prose original for the later versified epics.

Ibn al-Balkhī (wrote before 510/1116)

The author of the Fārsnāme is, strictly speaking, anonymous and the name Ibn al-Balkhī is conventional. The book was written for the Saljuqs and it contains much material on pre-Islamic Iran. Its sources include the books of Ḥamza al-Iṣfahānī and al-Ṭabarī (p. 8), but there is also a lot of material which is not known from other sources. However, Ibn al-Balkhī does not identify his sources when quoting from them. On p. 13, he mentions “histories and books about the genealogies of the Persians,” which probably refers to Arabic works. He also knows Ardashīr’s ʿuhūd o-waṣāyā (nuskhat’hā-ye ān mawjūd ast).

Mujmal al-tawārīkh (written in 520/1126)

The Mujmal is an anonymous work written in 520/1126 at the time of the Caliph al-Mustarshid, during the reign of Sanjar, son of Malikshāh, when Maḥmūd ibn Malikshāh was the crown prince.²³⁰ The author started his work earlier, on the instigation of a gentleman from Asadābād, but finished it only in 520 (Mujmal, p. 7/8–9).

The author lists an impressive number of sources (p. 2/2), starting with the prestigious al-Ṭabarī, whom, however, he only quotes occasionally. At the second place he mentions Firdawsī’s Shāhnāme, which he calls the original text (aṣlī), while other Persian texts are but branches (shuʿbahā). These branches he divides into verse and prose texts. Among the former are Garshāsbnāme (of Asadī Ṭūsī, cf. p. 2/3), Farāmarznāme, Akhbār-e Bahman, and Qiṣṣe-ye Kūsh-e Pīldandān; among the latter Abū l-Muʾayyad’s prose, such as the stories of Narīmān, Sām, and Kay Qubād, the stories of Luhrāsf, Āghush-e Wahādān²³¹ and Kay Shikan,²³² al-Ṭabarī (again), the Siyar al-mulūk by Ibn al-Muqaffaʿ, and Ḥamza’s “collection” (majmūʿe). Finally, he gives the list of Ḥamza, lifted from Ḥamza.

Al-Ṭabarī, Firdawsī, and Ḥamza are quoted throughout the pre-Islamic part of the book, Ḥamza particularly often without indication of source or only with reference to the ultimate source.²³³ A Sikandarnāme is referred to on p. 27/31, Siyar al-mulūk on pp. 28, 29, 52, 58, 65, 74, 75/32, 33, 63, 72, 81, 95, 96, Bīrūznāme/Fīrūznāme/Pīrūznāme on pp. 32 (twice), 54, 57, 64 (twice)/37 (twice), 66, 70, 79,²³⁴ 80, ʿAhd-e Ardashīr on p. 51/61, Kitāb (al)-Hamadān on pp. 46, 57/56, 70, and “the Bahmannāme, in the copy (nuskha) which Ḥakīm Īrānshān ibn Abī l-Khayr put into verse”²³⁵ on p. 73/92. The author refers in general to “an old book” (kitābī kuhan) on p. 55/67.²³⁶

The author also mentions, p. 74/94, several books which he (obviously wrongly) dates to the Parthian period (Kitāb Marwak,²³⁷ Kitāb Sindbād, Kitāb Yūsīfās,²³⁸ Kitāb Sīmās). From the period of Ardashīr-e Bābakān he mentions, p. 74/94, wisdom texts by Hurmizd-Āfarīd, Bihrūz, Burzmihr,²³⁹ and Īzad-dād that were translated into Arabic. He also mentions Mani’s Kitāb-e Ṣuwar,²⁴⁰ p. 74/94, and Kalīla o-Dimna (p. 75/96).

All the numerous quotations from the Sasanian historical text Kitāb al-Ṣuwar (quoted as Kitāb-e Ṣūrat-e pādishāhān-e Banī Sāsān, Kitāb al-Ṣuwar, Kitāb Ṣūrat, Ṣūrat-e Sāsā[niyā]n, Kitāb-e Ṣuwar, Kitāb-e Ṣūrat-e Āl-e Sāsān, pp. 29 (twice), 30, 32/33 (twice), 35, 37) seem to come through Ḥamza, as do the references to Ta⁠ʾrīkh mulūk al-Furs and Khudānāme on pp. 67–68/85. The author also quotes Ibn Qutayba’s Maʿārif (p. 58/71).²⁴¹ The quotations from Siyar al-mulūk need not come from Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag; at least the story of Balāsh and the daughter of the King of India (pp. 58–59/72) was hardly in Ibn al-Muqaffaʿ’s work (see Chapter 3.3).

It is also worth noting that when he comes to the last Yazdagird, the anonymous author suddenly takes al-Ṭabarī as his main source (p. 66/83). While al-Ṭabarī is occasionally used in earlier passages, it is possible that here his other sources had little to communicate, as the original Khwadāynāmag may not have taken the story to the end of the Sasanian Empire (see Chapter 6.2).

Mujmal, pp. 60–62/75–77, mentions several longer stories without narrating them: Qiṣṣe-ye Nūshzād (p. 61/75), the Testaments (waṣiyyat’hā) of Nūshirwān (p. 61/76), and the story of Bahrām Chūbīn(e) (pp. 62/76–77). Likewise, he mentions on pp. 73–74/92–95, several long stories, for which he gives various dates: Qiṣṣe-ye Shādbahr o-ʿAyn al-ḥayāt (p. 73/92), Qiṣṣe-ye Wāmiq o-ʿAdhrā’ (p. 73/93), Qiṣṣe-ye Shamʿūn (p. 73/93), Qiṣṣe-ye Ṣadūq o-Ṣādiq o-Salūm (p. 73/93), Qiṣṣe-ye Jirjīs (p. 74/93 – the last three on Christian history),²⁴² Qiṣṣe-ye Wīs o-Rāmīn (p. 74/94), and Qiṣṣe-ye Sharwīn o-Khwarrīn (p. 74/95). As can be seen from the list, some of these texts are known to have existed, or still exist, as independent books, while others are only found within larger compilations.

Many of these sources we know to have been in Arabic, and the syntax of the Mujmal in some cases implies that the Persian text goes back to an Arabic original.²⁴³ On the other hand, the Persian names of Kay Khusraw’s battles (pp. 41–42/48–49: razm-e Pashan, razm-e Kāmūs, razm-e duwāzdah rukh, razm-e buzurg) might imply a Persian source.²⁴⁴ Mujmal, p. 7/8, mentions that the author has translated some of his sources from Arabic into Persian “because that is the habit of speaking today.”

Mujmal, p. 10/11, also quotes the Avesta (Ābistā), claiming that the time between Gayōmard and the last Sasanian ruler, Yazdagird, was 4,182 years, a piece of information that obviously cannot come from the Avesta, but must derive from some later Pahlavi work. This, however, is only given as a piece of oral information based on the Avesta (pārsiyān az kitāb-e Ābistā (…) chunīn gūyand ke …). Later, p. 72/92, he again mentions the Avesta, giving several variant forms for the title.

Muḥammad Ṭūsī (wrote after 562/1166–7)

The last mentioned date in the text of ʿAjāʾib al-makhlūqāt is 562/1167 (p. 300).²⁴⁵ The book uses a lot of material familiar from the later nāmes (e.g., p. 441: Garshāsf in India), but little from Firdawsī’s Shāhnāme. However, the author highly respects Ḥasan-e (sic) Firdawsī of Ṭūs (p. 246).²⁴⁶ The book contains dozens (if not hundreds) of references to Alexander, largely familiar from the various versions of the Alexander Romance,²⁴⁷ and to Anūshirwān’s miraculous deeds and journeys. Afrīdūn, Ḍaḥhāk, Bahrām Chūbīn, and Balīnās also appear rather often and there are many similarities to the Arabian Nights, but very few sources are specifically identified. The stories about Persian national history often exhibit unique features not known from elsewhere.

3.7 The Contents of Ibn al-Muqaffaʿ’s Translation

All our sources agree that Ibn al-Muqaffaʿ translated the Khwadāynāmag into Arabic, but unfortunately few quote explicitly from this work. We do have a large number of quotations attributed to Ibn al-Muqaffaʿ, but, as we have seen in Chapter 3.4, Ibn al-Muqaffaʿ wrote a great number of works, many of which contained material relevant for Persian national history, so we are rarely in a position to ascertain whether a piece of information comes from Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag or some other work of his.

Ibn al-Muqaffaʿ’s translation is usually said to have been titled Siyar mulūk al-ʿajam or Siyar al-mulūk, sometimes also Khudhāynāme. We do find siyar al-mulūk often referred to, but usually without further identification. Here, there are two separate problems. First, it is often unclear whether the reference is to a book title or just to “the lives of Persian kings” in general. Second, even when it is clear that this has to be taken as a book title, there are a number of works that are referred to under this title. Several of the works on Ḥamza’s list (Chapter 3.1) bore this title,²⁴⁸ in addition to which there are individual Persian kings (such as Ardashīr and Khusraw Anūshirwān) to whom separate sīra works were dedicated.

In order to understand what Ibn al-Muqaffaʿ’s book might have contained, we have to be wary of falling into a vicious circle, first attributing various pieces of information to Ibn al-Muqaffaʿ without sufficient evidence and then proving their provenance from Ibn al-Muqaffaʿ’s work by showing that they fit our reconstruction. Instead, one should only include material that is explicitly attributed to Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag, either referred to as his Siyar al-mulūk or his Khudhāynāme.

Whereas there seems to be good reason to assume that the Pahlavi Khwadāynāmag was written in the sixth century and did not cover the end of the Sasanian Empire (cf. Chapter 6.2), there is an important passage in al-Maqdisī’s Badʾ v: 197, which is attributed to the Khudhāynāme (wa-fī Kitāb Khudhāynāme), and narrates the death scene of the last Yazdagird (651), itself inserted within a chapter on the Caliphate of ʿUthmān (Badʾ v: 194ff.). The passage is also connected with the Arab conquest. As there is no evidence whatsoever that al-Maqdisī would have been able to read Pahlavi, it is rather obvious that this comes from some Arabic translation, which also explains the interest in the conquest.

In the beginning of the story about Yazdagird, al-Maqdisī quotes Ibn al-Muqaffaʿ as his authority for the amount of gold Yazdagird had in his teasuries (Badʾ v: 195), though the Khudhāynāme is not mentioned there. Taken together, Badʾ v: 195 (Ibn al-Muqaffaʿ as authority) and the continuation of the story in v: 197 (the Khudhāynāme as authority) make it rather certain that the whole passage comes from Ibn al-Muqaffaʿ’s translation of the Khwadāynāmag. This shows that Ibn al-Muqaffaʿ continued the story until the demise of the Sasanian Empire.

Another reference in Tārīkh-e Sīstān, p. 56, to ʿAbdallāh ibn al-Muqaffaʿ and (his)²⁴⁹ Kitāb-e Siyar-e mulūk-e ʿajam, shows that the book contained the story of Alexander and Roxanne and the building of the town of Arak in Sistan. Such a detail finds its Sitz im Leben only if the contents of the Alexandre Romance were used for the book, which implies that Ibn al-Muqaffaʿ added a substantial story into the original Pahlavi text as it is improbable that the Sasanian chronicle contained any materials from the Alexander Romance (see Chapter 2.3).

In a highly problematic passage, Balʿamī, Tārīkh, pp. 4–5, refers to Shāhnāme-ye buzurg wherein pisar-e Muqaffaʿ counts the time from the expulsion of Adam from Paradise until “our Prophet” as 6,013 years. The passage continues with the identification of Gayōmard with Adam, but it is not clear whether this comes from the same source or not. The quotation is problematic because in another version of the book, the Tārīkhnāme i: 5, the Great Shāhnāme is attributed to Ḥamza and Ibn al-Muqaffaʿ is only quoted through it. Whatever the original form was, it is apparent that Ibn al-Muqaffaʿ was interested in synchronizations – as were all later authors²⁵⁰ – and his translation of the Khwadāynāmag most probably contained a number of such synchronizations added by him to the original text. The passage is discussed in more detail in Chapter 4.3.²⁵¹

Al-Dīnawarī, Akhbār, p. 9, quotes Ibn al-Muqaffaʿ’s critical attitude towards the synchronization of Jam with Solomon.²⁵² Whether the quotation ultimately comes from his translation of the Khwadāynāmag or not, it does yet again show that Ibn al-Muqaffaʿ was interested in synchronizing the various strands of history – he may in this case disagree with others, but his own system, too, is built on synchronization. It is obvious that all such synchronizations go back to the Islamic period, as a Sasanian chronicle was hardly interested in aligning Persian with Biblical history.²⁵³ This was of interest only for Christian and Muslim readership.

Muḥammad Ṭūsī, ʿAjāʾib, p. 240 (s.v. madīne-ye Shūsh) contains a note attributed to Ibn al-Muqaffaʿ (text: al-MQNʿ) which relates to the building of ancient walls. Although no book title is mentioned, such information on building activities is common in books that probably draw on the Khwadāynāmag and could well have been included in his translation.

Ibn al-Muqaffaʿ’s other major translation, that of Kalīla wa-Dimna, gives us some idea of how Ibn al-Muqaffaʿ worked as a translator. As de Blois (1990) has shown, Ibn al-Muqaffaʿ freely added new chapters to the book, so that only a part of the present Arabic Kalīla wa-Dimna goes back to the (lost) original Pahlavi text and there is no reason to assume that even those chapters which probably went back to the Pahlavi text are exact translations of the original text. As shown in Chapter 2.4, in the eighth century (and later) “translation” did not mean what it means in the 21st century.²⁵⁴