Entangled Religions 13.5 (2022) er.ceres.rub.de

Did the Āẕar Kaivānīs Know Zoroastrian Middle Persian Sources?

Kianoosh Rezania Ruhr-University Bochum, Germany

The Āẕar Kaivānīs, a syncretistic religious school in the sixteenth and seventeenth centuries, combined elements from Islam, Zoroastrianism, Buddhism and Ešrāqī philosophy. The Dasātīr, written by the first authority of the group, Āẕar Kaivān (943/1533–1028/1618), is a bilingual text. Its first language is an artificial encrypted language, represented as the language of heaven; the second is a specific form of New Persian, i.e., with few Arabic words. This article argues that Dasātīr’s author employed the Zoroastrian Zand as a model for the construction of his book. It moreover demonstrates the trace of some Middle Persian lexemes in it. Accordingly, it concludes that the Āẕar Kaivānīs were familiar with the Zoroastrian Middle Persian literature, if perhaps only superficially. The article also scrutinizes where and when contact occurred between Zoroastrianism and the Āẕar Kaivānī school. As a result, it discusses the Zoroastrian concept of secret language and the necessity of its translation and interpretation, which provided the Āẕar Kaivānīs with the possibility to include the notion of a secret book in their own system of thought.

Āẕar Kaivānī school, Dasātīr, Zoroastrianism, Zand, secrecy, Safavid-Mughal, religious contact



Āẕar Kaivānīs is a syncretistic religious school combining elements from Islam, Zoroastrianism, Buddhism and Ešrāqī philosophy; its major texts were composed in the sixteenth and seventeenth centuries. The texts name a certain Kaivān, probably from Estaḫr, near Shiraz, as founder of the school. According to the Dabestān-e maẕāheb (The School of Religious Teachings),1 a heresiographical work from the mid-seventeenth century (see below) whose author must have belonged to this school, Kaivān lived from 943/1533 to 1028/1618. He must have left his homeland for India under pressure resulting from the intolerant Safavid religious policy to enjoy the religious freedom of the Mughal empire, and settled in Patna, probably in the year 1001/1592–3, or at the end of the sixteenth century.2 The Dabestān-e maẕāheb (Āẓar Sāsānī 2010, 4r) refers to the school by various names: Īzadīyān, Yazdānīyān, Ābādīyān, Sepāsīyān, Āzādān, Sorūšān, Hūšīyān, Anūšagān, Āẕar-hūšangīyān and, last but not least, Āẕarīyān.3


The Dabestān-e maẕāheb presents a hagiographical biography of Āẕar Kaivān, making it difficult to attempt a historical contextualization of the founder’s activities. Given the strong syncretism of the Āẕar Kaivānī school, it is difficult not only to identify the origin of its ideas, but also to trace the religious contours of the school, i.e., to demarcate it from its neighboring religious groups and clearly define its ideas. One could even raise the question of whether the representation of the school in the Dabestān-e maẕāheb is a heresiographical4 categorization of the Dabestān-e maẕāheb’s author, an idealized depiction of the school, or a historical description. In contrast to their diverse content, Āẕar Kaivān’s texts feature a homogeneous form: They are written in Persian, the official language of Safavid Iran and Mughal India, and clearly strive to avoid Arabic words. The texts’ preoccupation with a ‘pure’ Persian language also caught the attention of nineteenth-century philologists;5 this fascination was short-lived, however, since later research proved that the word formations encountered in these texts are highly artificial and often do not follow Persian morphology. The scholarly disappointment reached its highest point in the investigations into a book which the Āẕar Kaivānīs represent as ‘heavenly’: the Dasātīr-e Āsmānī.


The title dasātīr-e āsmānī literally means ‘Heavenly Professors.’ Given the Āẕar Kaivānīs’ efforts to avoid Arabic words, it might come across as an accidental irony that the title of their heavenly book, dasātīr, is the Arabic plural of the Persian word dastūr. The book includes 16 chapters: the first 15 chapters are ascribed to 15 shats, or prophets, starting from Mahābād and ending with Zarathustra, Sāsān I. and Āẕar Sāsān. The text does not mention any of the prophets known from the Abrahamic traditions; instead, the prophets’ names derive from Iranian mythology, Zoroastrian cosmogony or anthropogony or, in other cases, they remain unknown. A chapter titled Pand-nāma-ye eskandar ‘Alexander’s Book of Advice’ is placed after the chapter Nāma-ye šat zartošt ‘Prophet Zarathustra’s Book.’ Alexander is not called a prophet in the Dasātīr, yet Zarathustra is quoted as saying that “No one can receive the meaning of my words as he [Alexander] did” (D, 222).


The Dasātīr is a bilingual text. Its first language is an artificial encrypted language; the second is a specific form of New Persian, i.e., one which includes few Arabic words. The Dabestān-e maẕāheb represents the pseudo-language of the Dasātīr as follows:


چند مجلد از آن زبانی بود که به هیچ زبانی فرودینیان نمی‌ماند و آن را زبان آسمانی نامیده.


Some volumes of that [scil. the Dasātīr] are/were in a language which does/did6 not resemble any language of the people of lower religions and that is called ‘the Language of Heaven’7.


The Dasātīr describes itself as a heavenly book sent by God to Mahābād, the first prophet of the Dasātīr. In the first decades after the discovery of the Dasātīr, scholars made valiant efforts to decipher this ‘language of Heaven.’ Once scholars understood that it was an invented language, interest in the Āẕar Kaivānī texts waned.


In a recent article, Daniel Sheffield (2014) made the case that the concept of heavenly language in the Āẕar Kaivānī school is directly connected to older notions of Horūfīya. The arguments he presents to support this hypothesis can be summarized as follows:



Whereas Sheffield’s hypothesis about the Hurufist influence on the concept of celestial language is plausible, it cannot, on its own, explain the construction of the Dasātīr-e Āsmānī as a whole. In this article, I would like to argue that the Āẕar Kaivānīs might have used the general paradigms of Horūfīya and Noqṭavīya, but employed the Zoroastrian Zand as a model for the construction of the Dasātīr. We know already that the Āẕar Kaivānīs were aware of the Zoroastrian New Persian literature, as the Dabestān-e maẕāheb explicitly shows. Furthermore, this article will show that they were familiar with the Zoroastrian Middle Persian literature as well, if perhaps only superficially. I will also show that the Āẕar Kaivānīs did not use the concept of secrecy in their encounter with Zoroastrianism in order to draw in-group and out-group distinctions. On the contrary, I argue that the Zoroastrian concept of secret language and the necessity of its translation and interpretation provided the Āẕar Kaivānīs with the possibility to include the notion of a secret book in their own system of thought.

Celestial Language, Translation and Commentary in the Dasātīr-e Āsmānī


This investigation begins with a straightforward analysis of the structure of the Dasātīr. In each chapter of the book, a phrase, or often a sentence, is rendered in the celestial language, followed by a Persian ‘translation’ of the phrase from the celestial language. Occasionally some sentences are added to the translation and are offered as the commentary on the original text. The celestial language is demarcated from its Persian translation by the number of the passage, which appears at the beginning of the phrase in the celestial language, and by the letter ت‌ (t; for tarğoma ‘translation’) at the beginning of the translation, as is illustrated, for instance, in the Haydarabad manuscript of the book. In this manuscript, the beginning of the commentary is marked with the letter ش (š; for šarḥ ‘commentary’). These signs, moreover, are written in this manuscript in red ink, whereas the texts in both languages are written in black.8 This striking structure did not escape the attention of the first editor of the text, although he regarded the New Persian text as an actual translation of the Dasātīr text in its ‘heavenly’ language. In the epilogue to this edition, Mulla Firuz b. Kaus writes:


باید دانست که زبان اصل صحایف منزله اصلا و قطعا مناسبت بزبان زند و پهلوی و دری بلکه بجمیع السنه مشهوره طوایف مختلفه این زمان ندارد و در عصر خسرو پرویز […] حضرت ساسان پنجم اینصحف را بزبان فرس در غایت سلامت و فصاحت و بلاغت […] ترجمه فرموده هر یک از آیات بینات که محتاج بزیادت شرح و بسط است بعد ترجمه الفاظ آیات شرحی واضح مرقوم تا طالبان را دریافت بسهولت میسر گردد (D., 306).


It should be known that the original language of the revealed books does not resemble the languages Zand, Pahlavi, Dari or even any famous language of the different contemporary people at all. In the era of Ḫosro Parvīz, Majesty Sāsān V. translated these books into Persian with the highest correctness, fluency and eloquence. For each verse that needed a commentary he wrote a clear commentary after its translation so that the students could easily apprehend it.


To provide an example for this text structure I render in the following the paragraphs 40-44 and 47-52 of the chapter Nāma-ye šat vaḫšūr yāsān (D, 97-9). To allow better visualization of the text structure, I have rendered the texts in the celestial language red, the translation black, and the optional commentary blue. The sign for the demarcation of the celestial language from the translation is replaced by an asterisk:


  1. زابهاش هروند اسپ و فه هروار دم له چایند * فروزهاش بسیار است و بشمار در نیایند


  1. فر هوشام لی هروار خمیده هزهیشام خر میم بهنام اسپ که سارام اورنگرامان و حمیدگان نوند شکار هواند * فرشتگان بی‌شمار آفریده از ایشان نخستین خرد نخست است که همه خردها و آفریدگان زیردست اویند


  1. نزم مانیستار که فدسمند اسپ ویودسرو نویشرامان است * پس روان سپهر برتر که بس بزرگست و سالار همه روانهاست


  1. تیرسریربد وهو فرسار سروسریرام اسپ * پس تنبد و او سالار همه تنهاست و تن‌بد نام سپهر برترین است


  1. سیامکان و هرنامگان و هرنامگانیان و شاورام و تاو رام سارام خمیده هواند هزهو فستام پم هیشام * آزادان و وارستگان و تنان و تنانیان و گوهرها و ناگوهرها همه آفریده اویند از او آفرین برایشان




  1. فه سام زمرپان وای * بنام مهربان خدای


  1. تشتاریدن رام برج ادهابیغی و سابیغی * پاک شدن دو گونه است امیغی و رواییی


  1. هابیغی منا درافه افرکنون له بر تن و هموزیدگیها سجردن * امیغی دل را به بدی نه بستن و نکوهیدگیها ستردن مانند خشم و کام از دل زدودن


  1. و سابیغی هانچیم دم پرکتار یاج هاسد سلودن * و رواییی آنچه در آشکار بد باشد زدودن چون اویژگی و ناپازی آشکاری


  1. وهیم تشتاریدن فه جریفترپامد * و این پاک شدن به آب یفتر باشد و یفتر آبیست که رنگ و بوی و مزه او نگشته بود و بدبوی نشده ورنه گلاب و مانند آن پاک‌ون ستوده جم است


  1. وجرسود دم کاو سمید * و آب کرد در خورد تنه و توش باید دانست که آب کرد آنرا گویند که تن و چیز بدان پاک شود و آن در خورد تن آمده پس در خورد تن پیل رودی و مردم را آنمایه که درو سراپا فرو شود و بهر پشه همینه


It is important to highlight at this point again that the celestial text of the Dasātīr is represented as the original text, and was considered as such in the nineteenth century scholarly research as well. As far as the genesis of the book is concerned, however, it is the Persian text, encrypted into an artificial language, which should be considered the original. Interestingly, one can find a reflection on the ‘original text’ and its translation in the Dasātīr itself. The 70th paragraph of Jī-āfrām’s Book of the Dasātīr reveals the language of the ‘original text’ and its translation as well as the necessity of translation for the purpose of accessibility:


New Persian ‘translation’:


سخن خدا و نامه خدا و فرشته خدا و فرسته خدا دانستی


It is worthy knowing the speech of God, the book of God, the angel of God, and the envoy of God.




سخن خدا نه بگلو و کام و زبان است و آن خواستی است و گفتی بی اینهمه که چون پرمود فرشته سالار بهمن بهستی پیوست و زین خامه بدست نیرو جهانرا نگاشت و یزدانی نامه دو است نامه نخستین دو گیتی است و آنرا مهین‌‌نامه گویند و بزبان فرازآباد فرز دساتیرش نامند که مهین‌نامه یزدان باشد و نامه دیگر دساتیریست که چم آنرا مه‌آباد و دیگر پیغمبران از مه‌آباد تا من یافته‌اند, و آن آرشی است که بر دل تابد نه باد نوا. و این باد نوا آنرا کالبد است بهر شنوانیدن و این را بفرتین نواد دریک دساتیر خوانند که کهین‌نامه یزدان باشد و مهین‌پیغمبرش خرد است […] و این فرز فرجیشور است بزبان دساتیر که بپارسی دری مهین‌پیغمبر باشد و دوم پیغمبر مردم است و او را انگیخته‌اند تا فرودیانرا بخواند (D., 68).


“The speech of God exists not by means of the throat, the mouth or the tongue: It is a will and a speech without any of these. For when He commanded, the chief of angels, Bahman, came into existence, and with this pen, he [i.e. Bahman] wrote the world with the hand of might. There are two divine books. The first book is the two worlds, and it is called The Great Book [mihīnnāma], and in the language of Farzābād, it is called the Farz-Dasātīr, that is, The Great Book of God. And there is another dasātīrī book, the meaning [chim] of which Mahābād and the other prophets from Mahābād down to me have acquired, and it is a signification [āriš] which shines on the heart, not [comprehended through] the breath of the voice. This breath of the voice is a mere from [kālbod] for it in order to make it heard [bahr-i shinavānīdan]. In the heavenly language [farātīn navād], it is called Darīk Dasātīr, which is The Small Book of God [kehīnnāma-ye yazdān]” (Sheffield 2014, 170). Its great convey is knowledge […]. This is called Farz-Farğīšvar in the language of the Dasātīr. This means ‘Great Envoy’ in Dari Persian, and designates the second envoy of people. He has been commissioned to call inferior people.”


The text structure of the Dasātīr, as shown in the above paragraphs, reveals three distinct components: the revelation to the prophets in a celestial language; the translation of the revelation; and finally, a commentary on the revelation. Both translation and commentary are represented as deriving from ancient times and are hence endowed with more value. As a result, not only the constructed celestial language is important for the composition of the Dasātīr, but also the artificial Persian language of the translation—from which words of Arabic origin are expunged.9 In my opinion, the systematically antiquated language of the translation and commentary are also an aspect of the author’s intention to present a ‘celestial language.’ The celestial message can only be received through prophetic mediation; therefore, divine action is expressed in the celestial text as well as in the translation and commentary of the prophetic figures. The purpose of the ‘pure’ language of the translation and commentary is not only to suggest their ancient origins, but also to allude to an idealistic past, namely the Sasanian period. In this way, their ancient character also confers authority on them.

Exegetical Traditions in the Āẕar Kaivānīs’ Environment


The most influential religious traditions in the Āẕar Kaivānīs’ milieu which possessed an exegetical tradition include the Vedic tradition, Zoroastrianism and Islam. For the sake of argument, I assume that the author of the Dasātīr was familiar with these exegetical traditions and might have used them as models for the construction of his ‘heavenly book.’


There is no doubt that the Āẕar Kaivānīs became familiar with the religious books of India after their migration to the subcontinent, if not even earlier; this is proven by the use of Sanskrit words in the Dasātīr as well as in other Āẕar Kaivānīs treatises. The following passage of the Dabestān-e maẕāheb, moreover, demonstrates the Āẕar Kaivānīs’ familiarity with the Vedas:


گویند کلام الهی آن است که هیچ یکی از آخشیجی پیکران بدان لغت متکلم نشوند و قرآن اگرچه کتاب آسمانی است اما تازیان را همان گفتار است و چهار بید که به زعم ایشان نامۀ سماوی است به لغت سنسکریت است که در هیچ شهری بدان زبان تکلم نکنند و سوای کتب این طایفه یافته نشود و گویند که این کلام فرشتگان است و بید کلام برهماست برای انتظام جهانیان.


They regard the celestial language as a language in which none of the elemental forms have been expressed. Although the Qurʾān is a divine revelation, the Arabs speak in its language. The four Vedas, however, which they consider a heavenly book, are in Sanskrit, a language not spoken in any region and found nowhere other than in the books of this group. They maintain that this [scil. celestial language] is the speech of angels, and that the Vedas are the speech of Brahmā for the arrangement of the worldly affairs.10


This passage might even give the impression that the author of the Dasātīr used the Vedas as a model for the construction of the celestial language in his heavenly book. It states that the Brahmans regard the Vedas as a heavenly book. This claim is justified with the argument that Arabic is the language of some people and therefore a terrestrial language, while Sanskrit, in contrast, is not a spoken language. Considering the existence of a commentary in the Dasātīr, a commentary on the Vedas could have served the model for the construction of Āẕar Kaivān’s heavenly book if Sanskrit had been used as a model for its celestial language. The Veda exegeses of Sāyaṇa ācārya, one of the most prominent intellectuals of medieval India,11 are considered the most important exegeses of the Vedas.12 He authored them at the height of Indian literature in the fourteenth century in the Vijayanagra Empire. Sāyana and his team penned 18 comprehensive exegeses on different Vedic works, which rapidly won authority. Their historical proximity to the Dasātīr’s creation, and their widespread reputation in India, allow us to assume that they were not unknown to the author of the Dasātīr. If he had aimed to construct his heavenly book modeled on a commentary on Vedic texts, it is logical to assume that he must have chosen a commentary by Sāyana, perhaps specifically the Ṛgvedasaṃhitābhāṣya,13 his commentary to the Ṛgveda. It should be noted, however, that this commentary—as virtually every other authoritative commentary on the Vedas—is written in Sanskrit. The original text and the commentary are thus written more or less in the same language, even if a speaker of Sanskrit cannot always understand a Vedic passage. Moreover, this commentary evidences a textual structure14 which definitely differs from one of the Dasātīr. In Ṛgvedasaṃhitābhāṣya the commentary encloses the commented text,15 whereas in the Dasātīr the commentary follows the original text.16

Figure 1: Fol. 31v of a Hs of Sāyaṇas Ṛgvedasamhitā­bhāṣyabhūmikā, RV I, 1.1 in center, surrounded by commentary , 296(Galewicz 2009, 296).


Commentators writing in the same language as the original text are not unique to the Vedas; this was true for some Qurʾān exegeses in Iran as well, where the most important commentaries were often written in Arabic. Commentaries with a Persian translation, however, were not infrequent in Iran. According to Zadeh (2012, 264–66), they linked the original and the translation in two forms: often through an interlinear translation, or by putting the translation at the end of a liturgical unit. The second form was not so current as the first one but common. The Persian translations of the Qurʾan thus incorporate three components similar to the Dasātīr: the original sacred text in Arabic, the translation, and the commentary in Persian:


Yet it is not uncommon for translations to fully envelop the text with the commentarial expansions. In these instances, the original Arabic text of the Qurʾan is not only contained between interlinear translations, above and below, but is also surrounded by marginal commentaries which fill the entire page so that the sacred scripture is visually afloat in a sea of exegetical expansion.17


As a consequence, it cannot be ruled out that Persian exegeses of the Qurʾan served as a model for the construction of the Āẕar Kaivānīs’ heavenly book. Nevertheless, there are some decisive differences between the Dasātīr and the exegeses of the Qurʾān or Vedas: in the commentary on Vedas, there are only two textual components, the original and its commentary. The Dasātīr has three components, however. In the Qurʾān, the original text is in a real, generally comprehensible human language, whereas in the Dasātīr, the original language is an artificial one. The texture constitutes the next major difference: The Ṛgvedasaṃhitābhāṣya, for example, exhibits a ring structure not present in the Dasātīr. In the case of the Persian commentaries on the Qurʾan, we frequently see an interlinear translation. Even when the translation appears at the end of a liturgical unit, the commentary, however, is often written on the margin. The commentary is thus not an integral part of the text as is the case for the Dasātīr. These differences make it improbable that these commentary traditions would have functioned as models for the Dasātīr.

The Zoroastrian Exegetical Tradition


In the second millennium CE, Zoroastrians, laity as well as religious specialists, believed that Avestan was a heavenly language. They regarded it as the language in which Zarathustra communicated with Ahura Mazdā. The knowledge that Avestan, as an Old Iranian language, had been spoken by a group of eastern Iranian people was promoted by Iranian philologists in the eighteenth and nineteenth century.18 Afterwards, Zoroastrians adopted this conclusion as well. Before these philological investigations, the general opinion did not consider Avestan to be a dead language but a language of revelation, not spoken by people on the earth. A thirteenth century Zoroastrian text adopts this perspective on the Avestan language:19


و آن بیست و یک نسک اوستا که میگویند اوستا زبان اورمزد است و زند زفان ما و پازند آنکه هر کسی +بداند که چه میگوید.


About those 21 nasks [scil. books] of the Avesta which they recite: Avestan is Ahura Mazdā’s language, Zand is our language and Pāzand is the one of which everybody knows what it [scil. Avesta] says.20


In the Zoroastrian tradition, Middle Persian (and its Pahlavi script) thus occupied an intermediate position between Avestan as an ideal language and New Persian (or Gujarati) as a spoken language. On the one hand, Middle Persian made the content of Avestan liturgical texts accessible to Zoroastrian priests; on the other hand, it historically stands at the interface between a language projected back into the mythical past and a living language.


The quoted passage, moreover, not only claims that Avestan is the language of Ahura Mazdā; it also introduces two other Zoroastrian linguistic components, Zand and Pāzand, which are relevant for our discussion. As we know, the Avestan texts were translated into Middle Persian and were commented upon.21 The commented translation written in the Pahlavi script is also known by the technical term zand, lit. ‘interpretation.’22 Since the complexity of the Pahlavi script hampers the reading of Pahlavi texts, some of these texts were re-rendered in the more distinctive Avestan script. So, the Middle Persian texts, occasional exegeses of Avestan texts, written in the Avestan script, are called pāzand. Therefore, we have to differentiate between the pair translation-commentary and zand-pāzand. The definitions of Zand and Pāzand in the quotation above are consistent with their definitions in Iranian philology (Andrés-Toledo 2015, 524). The quotation defines zand as ‘our language,’ i.e., the Middle Persian language, the literary language of the Zoroastrian priests in the Sasanian and early Islamic period, written in Pahlavi script, which in the period after the eleventh/twelfth century, in particular, Zoroastrian priests were able to read. In contrast, Pāzand is represented as a text form “of which everybody knows what it says.” The author presumably intends ‘everyone’ to mean lay Zoroastrians, who must have been able to read the Avestan script.


To illustrate the overall structure of the Zand texts,23 I will quote two verses of the Yasna text, Y. 9.1-2, from the exegetical tradition.24 These texts comprise, like the Dasātīr, three components: the original text, its translation, and the commentary. In the Zand texts, the Avestan passages are mainly translated phrase for phrase. In order to do this, first the Avestan original phrase is written (here rendered in red). Secondly, its translation follows (here rendered in black). Thirdly, a short or long commentary is occasionally added after the translation (here rendered in blue). In manuscripts, the original Avestan text is demarcated from the translation by a decorative character (here marked by an asterisk). Moreover, some words, such as hād, mark the beginning of the commentary.




hāuuanīm ā ratūm ā haomō upāit̰ zaraθuštrǝm * pad hāwan radīh [[pad hāwan gāh]] hōm abar raft ō Zardušt


* ātrǝm pairi yaoždaθǝṇtǝm gāθåsca srāuuaiiaṇtǝm * pad ātaxš-gāh pērāmōn yōǰdahrēnišnīh ka-š [[ān Ašem-wohū sē]] guft [[kē Frawarānēy ō pēš]]


* ādim pǝrǝsat̰ zaraθuštrō kō narǝ ahī * u-š az ōy pursīd Zardušt kū kē mard hē [[hād nē pad yašt ī fradom būd az pēš paydāg. u-š dānist kū hōm ōh rasēd ud ka mad būd ā-š pursīd abāyist mad miθrō upāit zardušt ān paydāg kū-š šnāxt ēd rāy čē ān zamān abāg yazdān wēš būd ēstād u-š yād āšnāgtar būd hēnd. u-š ēn fragard warm būd u-š abāyist rāy abāg hōm ul guft. * ast kē ēdōn gōwēd hād * Ohrmazd guft ēstād kū harw dō ōh rasēnd ud ka hōm mad būd ā-š madan šnāsēd.]]


* yim azǝm vīspahe aŋhǝ̄uš astuuatō sraēštǝm dādarǝsa xᵛahe gaiiehe xᵛanuuatō amǝṣ̌ahe * kē man az harwisp axw ī astōmand ā-m nēktar dīd hē čē-t ān ī xwēš ǰān nēk kard ēstēd ud amarg [[hād ā-š pad frārōnīh ā amarg kerd estēd nē ēdōn čiyōn awēšān kē gōšt ī ǰam ǰūd u-šān andar tan amarg kerd estād tā bē az tan harw kas-ēw amarg […]]]




āat̰ mē aēm paitiiaoxta haomō aṣ̌auua dūraošō * ō man ōy passox guft hōm ī ahlaw ī dūrōš [[hād dūrōšīh-iš ēd kū ōš az ruwān ī mardōmān dūr dārēd * rōšn guft ay ahōšīh pad hōm bawēd.]]


* azǝm ahmi zaraθuštra haomō aṣ̌auua dūraošō * an ham Zardušt hōm ī ahlaw ī dūrōš


* ā mąm yāsaŋᵛha spitama frā mąm hunuuaŋᵛha xᵛarǝtǝ̄e * ān ī ān ī man ōh ān xwarišn xwāhēd Spitāmān frāz man hūn ō* xwarišn [[xwarišn rāy bē hūn * xwarišn xward]]


* aoi mąm staomaine stūiδi yaθa māf aparacit̰ saošiiaṇtō stauuąn * abar man pad stāyišn stāy [[yazišn]] čiyōn man pas-iz sūdōmand stāyēnd [[ā-š ān ī tō ud tō ud ašmā rāy]]

Figure 2: Fol. 57v from Yasna Pahlavi Hs T55 (Andrés-Toledo 2012).


The migration of the Avestan texts from Eastern Iran to Western Iran, as well as some probable discontinuity in the Zoroastrian textual tradition, led to a situation in which the Zoroastrian priests of the post-Achaemenian period were not able to produce new texts in Avestan. It moreover undermined their competence in understanding the Avestan language. Due to these circumstances, translation of the Avestan texts became necessary and also increased the necessity for explanatory exegesis. Therefore, the Avestan original and its translation always accompany the exegeses. Consequently, Zand designates both the translation and the commentary of the Avestan text, although the Zoroastrian priests differentiated between them in their textual tradition. In the late or post-Sasanian period, the translation and the exegesis became fixed and acquired an authoritative status, which is partly projected in the Zoroastrian tradition on the Middle Persian language and the Pahlavi script. Whereas Avestan was considered Ahura Mazdā’s language, Pahlavi was represented as the language and the script of its mediators, that is, the Zoroastrian authorities. The 99th chapter of the Zoroastrian book Saddar-e nas̱r (Hundred Chapters in Prose), a Zoroastrian treatise from the fifteenth century or earlier, illustrates this Zoroastrian perception:


در نود و نهم. (1) اینکه موبدان و دستوران و ردان و هیربدان را نشاید که همه کس را پهلوی آموزند. (2) که زردشت از هورمزد پرسید که پهلوی آموختن مر کسان را شاید (3) هورمزد به افزونی جواب داد که هر که از نسل تو باشد موبد و دستور و هیربدی که خردمند باشد. (4) دیگر هیچ کس را نشاید جز از اینکه گفته‌ام اگر دیگران را آموزند او را عظیم گناه باشد اگر بسیار کارکرفه کرده باشد فرجام او را بدوزخ بود (Dhabhar 1909, 66).


Passage 99: (1) It is not allowed that mūbeds, dastūrs, radān and hīrbeds teach Pahlavi to everybody. (2) For Zarathustra asked Ahura Mazdā who is allowed to be taught Pahlavi. (3) Ahura Mazdā answered in detail, whoever is of your descendants (and) is a wise mūbed or dastūr or hīrbed. (4) Otherwise, nobody is allowed. If someone teaches someone other than those whom I have mentioned, s/he commits a huge sin. Even if s/he has many virtues s/he will be finally brought to hell.


Passage 99 limits instruction in the Pahlavi script and language to the Zoroastrian priests. It is worth noting that the restriction of teaching to priests refers only to the Pahlavi script and language. In contrast, Zoroastrians must learn the Avestan script to be able to accomplish their liturgical tasks, and priests must help them do so, as passage 98 of the same text requires:


در نود و هشتم (1) اینکه بهدینانرا می‌باید که خط اوستا بیاموزند پیش هیربدان و اوستادان تا در خواندن نیایش و یشت خطا نرود. (2) بیشتر واجب مر هیربدان را و اوستادان را هست که همه بهدینانرا خط اوستا بیاموزند و اگر هیربد در آموختن ایشان تقصیر نماید او را عظیم گناه باشد. (3) که اورمزد به افزونی زرتشت را گفت که هر هیربدی و اوستادی در آموختن اوستا بهدینانرا تقصیر کند او را از بهشت چندان دور کنم که پهنای زمین است.


Passage 98: (1) The Avestan script must be taught to Zoroastrians by hīrbeds and masters so that there will not be any mistakes in the recitation of prayers and Yašts. (2) It is more imperative to hīrbeds and masters to teach the Avestan scripts to all Zoroastrians. If a hīrbed neglects their teaching s/he commits a huge sin. (3) Ahura Mazdā emphasized to Zarathustra: ‘I will take every hīrbed and master who neglects teaching Avesta to Zoroastrians as far away from Paradise as the breadth of the earth.’25


Both passages attempt to authorize the presented direction through two postulates. The first postulate refers to the representation of the instruction as a divine provision, which was revealed to Zarathustra in a dialogue with Ahura Mazdā. The second postulate alludes to the representation of its violation as a severe sin, which leads the offender to hell even if s/he has acquired numerous virtues.


It is worth noting that these chapters are paraphrased in chapters 99 and 100 of the Dabestān-e maẕāheb:


در نود و نهم: بهدین باید که خط استا و ژند بداند.


در صدم: موبد را باید لغت پهلوی غیر را نیاموزاند چه یزدان به زردشت گفته این علم به فرزندان خود تعلیم کن.


Passage 99: Zoroastrians must know the Avestan and the Zand script.


Passage 100: Mūbeds must not teach Pahlavi words to others, because Yazdān [scil. Ahura Mazdā] has said to Zarathustra: ‘Teach this science to your children.’26


This demonstrates that this emic perspective on Zoroastrian exegetical literature was known to the Āẕar Kaivānīs, as the section on the reception of Zoroastrian exegetical tradition below will attempt to investigate in more detail.

Comparing the Structures of Exegetical Texts in Zoroastrianism and in the Dasātīr


The evidence presented above allows us to infer that, even if the Āẕar Kaivānīs took over the concept of celestial language from their immediate religious environment (Horūfīya and Noqṭavīya), their construction of the Dasātīr-e Āsmānī obviously imitates the Zoroastrian Zand. This hypothesis is supported by the following evidence:



If we accept that the author of the Dasātīr used the Zoroastrian Zand tradition as a model for his book, there would be no doubt that the Zoroastrian exegetical texts were known to the Āẕar Kaivānīs at the latest after their migration to India. Now the question can be posed to what extent these texts were known in the broader context of early Modern Iran and India and how deeply Āẕar Kaivānīs authors were acquainted with them.

Reception of the Zoroastrian Exegetical Tradition in Early Modern Indo-Iranian Culture and in Āẕar Kaivānī Literature

In Early Modern Indo-Iranian Culture


In the early modern period, Middle Persian was considered the language of the golden age of Iran and was often contrasted with contemporary spoken languages. Its importance was not restricted to Zoroastrianism; it was generally perceived as the language of pre-Islamic heritage. This is the case with philologists such as Ğamāl al-dīn Enğū Šīrāzī, the author of the famous Farhang-e Ğahāngīrī, composed between 1595 and 1608.28 His interest in Pahlavi philology must have been so great that at the end of the sixteenth century, Akbar (1556–1605), the third Mughal emperor, invited Ardašīr, a knowledgeable Zoroastrian priest from Kerman, to his court to help the philologist with his dictionary.29 As an epilogue to the lemma ‘barsam,’ thin branches of tamarix or pomegranate tree, which are used in Zoroastrian rituals, Enğū Šīrāzī writes:


شرح این لغت را مجوسی که در دین خود بغایت فاضل بود، و اردشیر نام داشت، و او را مجوسان موبد میدانستند و حضرت عرش آستانی محض، بجهت تحقیق لغت فرس مبلغها از برایش فرستاده از کرمان طلبیده بود، تحقیق نموده، نوشت (Enğū Šīrāzī [1351] 1972, I/854).


A Zoroastrian who was extremely learned in his religion, named Ardašīr, whom the Zoroastrians considered mūbed, and to whom the Majesty of the absolute empyrean throne sent an enormous sum of money, inviting him from Kerman for philological investigations of Persian, did some research and wrote the explanation of this term.


Ardašīr seems to be alluded to in the entry āẕar as well (Modi 1903, 90–91):


و فقیر حقیر که راقم این حروفم، پیری از پارسیان را که در دین زرتشت بود دیدم، که جزوی چند از کتاب زند و وستا داشت چون مرا رغبت و شعف تمام،‌ بجمع لغات فرس بود [–] و در فرس از زند و وستا کتابی معتبرتر نیست – بجهت تحقیق لغات با او صحبت میداشتم و اکثر لغاتی که در خاتمه‌ٔ کتاب از زندو وستا نقل شده، تقریر [آن] زرتشتی است (Enğū Šīrāzī [1351] 1972, I/96).


I, the little poor (man) who is the writer of these letters, saw a wise man of Persians/Parsis who was Zoroastrian. He had many parts of the book Zand-Avesta. As I was very interested in compiling Persian words and there is no more creditable book than the Zand-Avesta in Persian, I engaged in conversation with him because of (my) philological investigations. Most of the words that are listed at the end of the book of the Zand-Avesta are written by that Zoroastrian.


For our discussion, it is worth examining how the Zoroastrian terms zand, pāzand, and avestā were perceived in non-Zoroastrian environments in the early modern era. For this, I quote their definitions in the Farhang-e Ğahāngīrī and the Farhang-e Moʾaiyad al-Fożalāʾ:30


ابستا با اول مفتوح و ثانی مکسور، تفسیر زند باشد. و زند کتاب زرتشت است


(Enğū Šīrāzī [1351] 1972, I/563)


Avesta: [abestā] is the commentary on Zand, and Zand is Zarathustra’s book.


پازند تفسیر زند باشد، و زند کتاب زرتشت است (Enğū Šīrāzī [1351] 1972, I/231)


Pāzand: is the commentary on Zand, and Zand is Zarathustra’s book.


زنداستا بالفتح نام کتابی در احکام آتش‌پرستی از مصنفات ابراهیم زرتشت (Dehlavī, n.d., 432)


Zanda(ve)sta: the name of a book comprising instructions about fire-worshiping, of Ebrāhīm-Zardošt’s compositions.


زند […] نام کتابی از جمله مصنفات ابراهیم زرتشت در احکام دین باطل آتش‌پرستی که شرح پازندست […]


Zanda(ve)sta: the name of one of Ebrāhīm-Zardošt’s compositions comprising instructions of the false religion of fire-worshiping. It is the commentary on Pāzand.31


One can distinguish between the emic Zoroastrian definition of the terms Avesta, zand and pāzand, on the one hand, and their understanding in the broader milieu of early modern Indo-Iranian culture on the other. It appears that the author has mixed Avesta and Zand with each other: he represents Avesta not as the original but as the commentary, and Zand as the original text, whereas in Zoroastrian use it designates the commentary. The distinction between the original text and the commentary, however, is known to the author. The component translation is completely absent.

In Āẕar Kaivānī Literature


The chapter ‘On Some Benefits of Secrets of Zoroastrians’ (dar ẕekr-e baʿżī az favāyed-e romūz-e zardoštīyān) in the Dabestān-e maẕāheb describes the inaccessibility of revelation, the necessity of commentary and the division of commentary into two types, main and secondary:


بدان بعضی از یزدانیان گفته‌اند که کتاب ژند بر دو قسم بود: یک قسم آن صریح و بیرمز، که آن را مِه‌ژند نیز میگفتند، و قسم دوم رمز و اشارات که آن را کِه‌ژند هم میخواندند، و مه‌ژند مشتمل بود بر احیای شریعت حضرت مه‌آباد، چنانکه کتب آذرساسانیان، است، و مه‌ژند از تسلط بیگانگان، چون ترکان، خاصّه رومیان، از میان رفت و که‌ژند ماند، و بسیاری از که‌ژند هم در تاختها از میان رفت. خلاصۀ مضامین مه‌ژند آنکه […] در مطالب دیگر از علمی و عملی چون حفظ زندبار و قتل تندبار با دساتیر موافق است و در عهد اشکانیان عمل به که‌ژند کردند، چون اردشیر، مطیع ساسان دوم شد، عمل به دساتیر و مه‌ژند نمود و از قتل زندبار دوری جست، و مه‌ژند نیز جزو دساتیر است، و بعد از آن، دیگران رو به عمل که‌ژند آوردند، و انوشیروان بنابر اشارۀ آذرساسان عصر، عمل بر دساتیر و مه‌ژند کرده، از قتل زندبار مبرّا زیست، و باز بعد از او عمل به احکام که‌ژند کرده، تا ساسان پنجم نفرین در حقّ ایرانیان کردند و ایشان گرفتار فقر و مسکنت گشتند.


Know that some of Yazdānīyān have said that the book Žand comprised two sorts (of žand): one sort was unequivocal and without enigma, also called Meh-žand [the Higher Žand]; the second one included enigmas and allusions, also called Keh-žand [the Lower Žand]. The Meh-žand, like the books of the Āẕar-sāsānīds, contained the law of the holy Mahābād. The Meh-žand was lost during foreign conquests, such as those of the Turks and especially the Greeks. The Keh-žand, however, still remained, but a great part of it was also lost during invasions. In summary, the Meh-žand’s contents are […] In other matters, scientific and practical, e.g., the protection of harmless animals and killing of harmful ones, it agrees with the Dasātīr. In the Arsacid period, the people acted according to the Keh-žand. Ardašīr, obeying Sāsān II, acted according to the Dasātīr and the Meh-žand. Consequently, he avoided killing harmless animals. The Meh-žand is a part of the Dasātīr. After him, others began to adopt the Keh-žand, Following the contemporary Āẕar-sāsān’s authority, Anūšīrvān adopted the Dasātīr and the Meh-žand. Thus, he refused to kill harmless animals. After him, people again adopted the Keh-žand’s precepts until Sāsān V execrated Iranians and they fell victim to wretchedness and poverty.32


This passage illustrates that the Zoroastrian division of the texts into divine revelation, translation and commentary was not unknown to the Āẕar Kaivānīs. The artificially Persianized word žand, in particular, reveals that the author is working with the Zoroastrian concept of zand. I do not, however, claim that Meh-žand and Keh-žānd, as described in the passage, would coincide with the pair zand-pāzand or translation-commentary. Nevertheless, it seems plausible to assume that the Āẕar Kaivānīs were familiar with the Zoroastrian distinction between translation and commentary, which are together called zand: the author could thus have designated translation, which may still contain ambiguities, keh-žand, and interpretation, which explains the uncertainties of the translation, Meh-žand.


It is well known that the Āẕar Kaivānīs received some New Persian Zoroastrian works.33 This can be seen, for example, in the Dabestān-e maẕāheb, where the author explains the belief system of the Zoroastrians:34 there, some sections from works Zarādošt nāma,35 Ardā-vīrāf nāma,36 and Ṣaddar37 are paraphrased. This demonstrates that the Āẕar Kaivānīs were familiar, at the very least, with the New Persian literature of the Zoroastrians. In addition, the Zoroastrian priests directly participated in the inter-religious discussions at the Akbar court (see below). This likely added to the reputation of Zoroastrianism in this period, so that the Āẕar Kaivānīs might have been eager to know more about it after their arrival on the Indian subcontinent and might have attempted to come into contact with Zoroastrian priests. The author of the Dabestān-e maẕāheb, for example, claims to have been in contact with a Zoroastrian priest from Navsari:


در میان مردم مشهور است که زردشت آذربایگانی است، امّا غیر بهدینان گویند و نامه‌نگار از موبد برزو – که نوساری من اعمال گجرات وطن اوست – شنیده که مولد زردشت و آباء نامدارش شهر ری است.


It is common among the people to believe that Zarathustra comes from Āẕarbāygān. This however is what non-Zoroastrians say. The author has heard from mūbed Borzū, who is from Navsārī in the province Gujarat, that the birthplace of Zarathustra and his distinguished ancestors is the city of Ray.38


The author of the Dabestān-e maẕāheb even sets the religion of Zarathustra and the one of the Āẕar Kaivānīs in an exegetical relationship and claims that the former was adapted to the latter by interpretation, since the words of Zarathustra were mysterious:


چون این دانسته شد، بدان که کیش آذرهوشنگیان یعنی یزدانیان آن است که اگرچه دین زردشت از گشتاسپ تا یزدگرد رواجی تمام داشت اما تأویل کرده آن را با شریعت آذرهوشنگ یعنی مه‌آباد مطابق می‌ساختند و هیچگونه به قتل زندبار فرمان ندادند و کلمات زردشت را مرموز می‌دانستند جایی که مخالف کیش آذرهوشنگ بود عمل نمی‌کردند و تأویل می‌نمودند. […] آذرساسانیان جز به راه شت مه‌آباد نمی‌رفتند و کیشی دیگر بی‌تأویل نمی‌پسندیدند و اصلا ملتفت به ظاهر قول زردشت نبودند و ایشان برآنند که عقیدۀ خسروان خاصه دارا و داراب و بهمن و اسفندیار و گشتاسپ و لهراسپ بر این بوده یعنی کلام زردشت را حق می‌دانستند، اما ظاهر کتاب او را مرموز می‌شمردند.


Now that you understood these (premises), you should also know that the teaching of the Āẕar-hūšangīyāns, i.e., the Yazdānīyāns, states that although Zarathustra’s religion flourished from the time of Goštāsp to that of Yazdegird, they interpreted it and adapted it to the teaching of Āẕar-hūšang, i.e., Mahābād. They never recommended the killing of harmless animals. They considered Zarathustra’s words ambiguous and did not follow them when they contradicted Āẕar-hūšang’s teaching, instead reinterpreted them. […] The Āẕar-sāsānīs followed only the way of the prophet Mahābād. They did not accept any other teaching without interpretation, and did not adhere to the external form of Zarathustra’s words at all. They moreover believed that this was the opinion of (ancient) kings, especially Dārā, Dārāb, Bahman, Esfandīyār, Goštāsp and Lohrāsp. They accepted Zarathustra’s teachings as true but considered the exoteric aspect of his book symbolic [rather than literally true].39


Significantly, the author of the Dabestān-e maẕāheb claims that Bahrām b. Farhād Esfandīyār Pārsī, the author of the Šārestān-e čahār čaman, who died in 1624, knew Pahlavi:


فرزانه بهرام بن فرهاد از نژاد گودرز کشواد بوده. چون آذرکیوان به پنته خرامید در باز پسین روزها فرزانه بهرام از شیراز آمده در پتنه به ریاضت مشغول شد و او مردی بود مراتب منطقیات و طبیعیات و ریاضیات و الهیات از پارسی و پهلوی و تازی زبان آنچه نقل افتاده.


“Farzāna Bahrām the son of Farhād was from the lineage of Gūdarz, the son of Kashvād [an ancient hero from the Book of Kings]. When Āzar Kaivān went to Patna in his later days, Farzāna Bahrām came from Shiraz. He occupied himself with austerities in Patna. He was a man who had obtained the highest degrees and accolades, and he was well read in the sciences of logic (manṭeqīyāt), natural sciences (ṭabīʿīyāt) and theology (elāhīyāt) as transmitted through the Persian, Pahlavi, and Arabic languages.”40


These passages evince that the Āẕar Kaivānīs were familiar with the general concepts of the Zoroastrian commentary tradition. Moreover, they presumably were in contact with Zoroastrian priests who knew Middle Persian. We can thus search for the linguistic traces of contact with the Zoroastrian Middle Persian in the Āẕar Kaivānī texts, and particularly in the Dasātīr.

Some Pahlavi Terms in the Dasātīr


In the previous sections, I investigated the structural analogy of the construct Dasātīr and the Zoroastrian Zand tradition of the Avestan texts. I tried to demonstrate the Dasātīr’s structural dependence on the Zand tradition. Moreover, I tried to infer from the Āẕar Kaivānī literature that these authors were familiar with the Zoroastrian text tradition and knew Zand and its structure. In the following I would like to point out some terms in the Dasātīr that must have found their way to the Dasātīr from Zoroastrian Middle Persian literature. For this, I will concentrate on terms related to the concept of time. For my conclusions in this part, I formulate two explicit premises:



Both premises seem probable enough to be accepted as true and presupposed in the following. The first terms to scrutinize come from the commentary on section 29 of the chapter Šāy Kelīyo in the Dasātīr. There, we find two terms representing time which could be revealing for identifying the sources of the Dasātīr. The section reads:


میلاد ور ورد * آفریننده و پیداکننده بیمایه و دمانکش همه باید دانست دمان چندی گردش آسمان بزرگست و خویشی ناپاینده و نادرست بناپاینده و نادرست چون نو پدید امده و تازه شدهای روزانی را خویشی بگردش اسمانها و چرخ سپهران و این را بفراتین نواد زروان گویند (D., 78).


mīlād var vard * The creator and revealer is completely immaterial and without duration [damān-keš]42 It should be known that time [damān] is the measure of the rotations of the great sky, “and the relation of one fleeting and unfixed subject with another fleeting and unfixed subject; as for example, the relation of new events and fresh occurrences in the world, with the revolution of the Heavens and the motion of the spheres.”43 In the celestial language [farātīn-navād], it is called zorvān.


Striking in this passage is the word form damān, in damān-keš, instead of the New Persian word zamān ‘time.’ One might think this is a mere spelling mistake, where the letter <z> was replaced with <d> in the Persian-Arabic script. Although this confusion cannot be ruled out, it is hardly likely because of its repetition in different parts of the book. Much more likely is a misreading of a text in the Pahlavi script: In Pahlavi, the word zamān is written in two ways: <zmʾn’> or <dmʾn’>, where <d> is the corrupted form of the letter <z> (hence transliterated as <ẕ>). It is worth pointing out that before modern philological investigations, Zoroastrian priests read the word as damān. The use of the letter <d/y/g>44 instead of <z> is a well-attested phenomenon in the Pahlavi script, as the following Middle Persian words demonstrate:

<ẕmyk> as well as <zmyk> for zamīg ‘earth’

<ẕmstʾn’> as well as <zmstʾn’> zamestān ‘winter’

<yẕdʾn’> yazdān ‘gods’

<ʾwhrmẕd> ohrmazd ‘Ohrmazd’


The word form damān appears in other passages in the Dasātīr as well, where its meaning ‘time’ is explicitly confirmed:


دمانی چیزیرا گویند که هست نتواند شد جز در دمان که چندی گردش برترین سپهرست و هستی خردان باز بسته بدمان نیست و خرد نخست را کمان بدمان بود کردن چرخه آورد چه دمان برین نیرویش باز بسته بر سپهر باشد و هستی سپهر باز بسته بر هستی نخستین خرد (D., 256).


Temporal [damānī] is called that which can be created only in time [damān], which is the measure of the rotations of the greatest firmament. The existence of Intelligences does not depend on time [damān]. Making the First Intelligence dependent on an existence in time [damān] produces circular reasoning because time itself depends on the firmament for this (form) of its force, and the existence of firmament itself depends on the existence of the First Intelligence.


و هستی نزد دانش او یک‌بار بی دمان و هنگام پیداست و بر او هیچ چیز پوشیده نیست رسا دانایی که دانش او هنگامی نیست و در فر باره او گذشته و اکنون و آینده نگارش نتوان کرد کشش دمان و درازی هنگام بانو شدها که پیوسته لختان و لخت‌های اوست یکبار نزد یزدان پدیدار است (D., 3).


And the existence is manifest to His knowledge at once, without time [damān] and duration [hengām]; and nothing is hidden to Him. His knowledge is expressive because His knowledge does not have duration. It is impossible to ascribe to Him past, present and future. The progress of time [damān] and the length of duration, with renovations, which occur in continuous divisions, which are its [scil. time’s] divisions, are manifest to God at once.


و اخترشناس +خروس است که دمان و هنگام روز و شب نیکو شناسد (D., 213).


The cock is an astronomer who knows time [damān] and the duration [hengām] of the day and night right well.


یکتای بی ‌امید مزد از بخشندگی و نیکویی کردن نخست آزاد و رسته گوهری بی‌پیوند و بند و مایه و پیکر و دمان و هنگام و تن و تنانی و نیاز و آرزو به تن […] آفرید (D., 4).


Without hope of return, only for generosity and beneficence, the unique One, first of all, created an essence free and unlimited, independent, boundless, immaterial, formless, timeless [()-damān], without duration [()-hengām], without body and bodiness, without need and wish to body […]


The use of the word form damān instead of the New Persian zamān in the quoted passages from the Dasātīr can be explained with one of the following reasons:



Since in the sixteenth century only the Zoroastrian priests had the competence to read the Pahlavi script, one is forced to conclude from this word form that either the author belonged to this circle, which current scholarship does not support, or obtained his information from Zoroastrian priests. In any case, he must have used a Pahlavi text as a source, directly or indirectly.


Decisive is likewise the time term used in the celestial language (farātīn navād),45 zorvān. The word derives from MP zurwān, which in turn is a loan word from Avestan zruuan- ‘time,’ and appears as a New Persian word only in the Zoroastrian literature. In the sixteenth century, the name could have been derived from a Pahlavi text, an Arabic work of heresiography such as al-Šahrestānī’s al-Milal wa-l-niḥal, its translation into New Persian, or a New Persian Zoroastrian account of the Zurwān myth.46


The only New Persian treatises known in the scholarship that deal with the Zurwān myth or the Zoroastrian theory of time are ʿOlamā-ye eslām (UI), ʿOlamā-ye eslām be dīgar raveš (UIbdR) and a short passage quoted below. The word zorvān, however, does not appear in these works; to denote profane time, UIbdR uses zamān, zamān-e derang-ḫodāy (mp. zamān ī dagrand-xwadāy) (UIbdR, 81.13) or zamāne (UIbdR, 84.8); for the designation of the sacred time, it uses zamān (UIbdR, 81.6-9, 82.16) and zamāne (UIbdR, 82.16,18). Similarly, UI uses zamān, zamāne and rūz(e)gār to denote profane time.47 In another New Persian passage,48 which alludes to the Zurwān cosmogony, sacred time is again referred to as zamāne. In other New Persian Zoroastrian accounts that the Āẕar Kaivānīs received, such as Zarātošt-nāme, Ardā vīrāf nāme and Ṣaddar, the word zorvān—as far as I discovered—does not occur. Therefore, the word zorvān could not have been taken from these New Persian Zoroastrian works in the mentioned section from the Dasātīr.


Some Arabic heresiographies deal with the Zurwān myth, especially the al-Šahrestānī’s al-Milal wa-l-niḥal. It is obvious that the Āẕar Kaivānīs knew and received al-Šahrestānī’s book. The Dasātīr even contains direct quotations from the Arabic original, and not its New Persian translation.49 Therefore we are tempted, at first glance, to assume that Āẕar Kaivān adopted the word zorvān from Šahrestānī’s book. A more attentive examination of the text passages in question, however, shows that zorvān does not have the meaning ‘time’ in these passages.50 There, zurwān is only presented as a primordial principle; the word does not represent a concept of time or eternity. This is true also for other Arabic heresiographies that narrate the Zurwān cosmogony.51 In some descriptions of Zoroastrianism in the Dabestān-e maẕāheb, one can recognize Zurvanite traits. None of these sections, nonetheless, indicates that the author used the word zorvān or azorvān to mean ‘time; eternity.’ These passages are listed below:


بهدینان گویند زردشت شاخی از بهشت آورده، بر در کشمیر نشاند و این سَروْ شد و نزد یزدانیان این سخن اشارت است بدان که نَفْس مجّرد در نبات است و بعضی از یزدانیان گفته‌اند زردشت از ربّ سَروْها که آن را ازروان گویند درخواست تا کِشته او را نیکو پرورد و از یکی از حکمای مرتاض نقل کنند که گفت من رَبّ سَروْ را دیدم، فرمود که من متوکّل را کُشتن فرمودم، به جُرم بریدن آن.


The Zoroastrians believe that Zarathustra brought a branch from paradise and planted it at the gate of Kashmir; this grew up into a cypress. According to Yazdānīyān, this saying alludes to the fact that the incorporeal soul is vegetable. Some Yazdānīyāns narrate that Zarathustra asked the lord of cypresses, who is called Azarvān, to carefully nourish this (tree) that he had planted. They narrate the following from one of the ascetic savants: “I saw the lord of cypress, and he commanded: ‘I ordered that Motevakkel be slain for the crime of cutting that cypress.’”52


اکنون هنگام آن است که لختی از رمز و اشارات شت زرتشت را آورد چه از رمز حکمت محفوظ ماند و به دست نابخرد نیفتد و کامل مطلب ازو برگیردبالجمله پیروان زردشت گویند گیتی را دو صانع است، یزدان و آهرمن یزدان اندیشه بد کرد و گفت که مبادا مرا ضدی پدید شود که دشمن من باشد آهرمن از فکر او پدید آمد و در بعضی جا آمده که ایزد تنها بود آن را وحشتی پیدا شده فکر بد کرد، آهرمن پیدا گشت […]


It is now time to present some of the enigmas and allusions of the prophet Zarathustra, as enigma guards wisdom from falling into the hands of ignorant, and only perfect ones can benefit from its content. For example, Zarathustra’s adherents believe in two creators of the world: Yazdān and Āherman. Yazdān conceived an evil thought and uttered: “Perhaps, an antagonist may arise against me who shall be my enemy.” Āherman arose from this thought of him. Otherwise, it is attested in some places that Yazdān was alone, a fear overwhelmed him, he had an evil thought and Āherman arose.53


بهدینان گویند: «آهرمن از زمان پدید آمد، و فرشته‌ها و آسمانها و ستارگان بودند و باشند، اما پدید آمدۀ موالیدند، و مدّت ماندن این آفرینش دوازده هزار سال است، پس رستخیز شود و یزدان مردم را برانگیزد و همین جهان آخشیجی را بهشت برین سازد و آهرمن و آهرمنان و دوزخ را به نیستی برد».


The Zoroastrians believe that Āherman arose from time, and that the angels, skies and stars existed and will exist, but are the result of births. The period of this creation is twelve thousand years. Afterwards, the resurrection will occur. Yazdān will resurrect the people and transform this material world into the eternal paradise. He will annihilate Āherman, his adherents and hell.54


The word zorvān is not used in the time theory of the Āẕar Kaivānīs as described by the Dabestān-e maẕāheb,55 although pseudo-words are artificially constructed to designate different time periods of the multi-period world age. These periods and their relations are shown in the following table:

zād vād ğād mard vard fard sāl māh rūz
world age 100 2.16 × 1026
zād 2000
vād 3000
ğād 1000
mard 1000
vard 1000
fard 106
sāl (‘year’) 12
māh (‘month’) 30


Consequently, no other literature remains except Zoroastrian Pahlavi literature to serve in the quoted section of the Dasātīr as a source for the use of the word zorvān. Accordingly, the author of the Dasātīr must have taken the two words for time, damān and zorvān, from the Zoroastrian Middle Persian literature, directly or indirectly through the Zoroastrian priests. The assertion that in the celestial language ‘time’ means zorvān is also decisive for the following reason: it explicitly shows that for Āẕar Kaivān the template for the celestial language was the Avestan language, in which the word zruuan means ‘time.’ Dasātīr’s designation of the celestial language, farātīn-navād, mentioned in the quotation above, occurs in three places in the book (D., 69, 78 and 263). Besides the passage quoted above, the following passage is significant for identifying the template of the celestial language:


چنانکه همی نموده آمد انرا گوهر خوانند و بفراتین نواد فروهر (D., 263).


As it has been shown, it is called essence, and in the celestial language [farātīn-navād] fravahr/frūhar.


The author here again uses a Zoroastrian terminus technicus, which derives from Avestan (< frauuaṣ̌i-), as a celestial term. This usage increases the probability that the Dasātīr’s author designed his book after Zoroastrian Zand texts, with Avestan in mind as a template for his celestial language.

The ‘Where’ and ‘When’ of the Religious Contact


The historical contextualization of Āẕar Kaivān’s encounter with Zoroastrianism faces many difficulties, and this is true even for the historical contextualization of the school itself. When did Aẕar Kaivān live? And when did he migrate to Patna? Who authored the Dasātīr, and when? Even these most basic questions can be answered only tentatively because we have only late manuscripts of the Āẕar Kaivānī texts at our disposal. The same questions can be raised regarding the Dabestān-e maẕāheb, a text whose authorship has been the subject of controversial discussion. The discovery of an old Dabestān manuscript, however, contributes enormously to answering some of these questions.


Some years ago, the Cultural Center of Iran in New Delhi acquired a Dabestān manuscript dated to 8 Shawwāl 1060 H. (1650 A.D.). The colophon of the manuscript reads:


کاتبه العبد فقیر حقیر محمد شریف ابن شیخ میان سپاهی زاده بوم میدک ساکن بنده تبلحور تحریراً فی التاریخ هشت ماه شوال سنه۱۰۶۰. تمت تمام شد، شیطان غلام شد.


Written by poor, abject Muḥammad Šarīf b. Šayḫ Mīyān, soldier, born in the land Mīdak, resident of Banda-ye Tabalhūr (?), recorded in the date, 8, month Šavvāl, year 1060 [October 4 1650]. finitur, completed, Satan became slave.56


This makes it the oldest known Āẕar Kaivān manuscript, 15 years older than the Mashkut manuscript of the Nāme-ye zardošt or Zūre-ye bāstānī. The most salient feature of this manuscript is that, on the 23 Shawwāl of the same year, a student of the author compared this manuscript with what was apparently the original text of the author and noted the differences on the margin of the manuscript. He records his activity in an epilogue to the manuscript as follows:


بانجام انجامید مقابله دوازده تعلیم از کتاب دبستان که انشای مرشد المحققین، امام المدققین، عارف کامل، صوفی واصل، حکیم حکمت‌کده دریافت حق، شناسنده معارف حضرت وجود مطلق، مؤیّْد بتأییدات سبحانی، اعظم شافی، استاد میرزا ذوالفقار آذر ساسانی المتخلص بموبد، طول الله عمره، که بسنه ۱۰۶۰ بسلک تألیف درآمده بود، بقدر طاقت صحت داده و قید و قیود و بتوان شناخت خود کرده در کناره ثبت نموده بنشان میم. امید که از خطا در امان خدا باشد. انشاءالله آنچه دیگر تألیف شود نگاشته گردد.کمین شاگرد مجدالدین محمد، مقابله‌ساز این نامی‌نامه حضرت استاد است. قد حرّر فی ۲۳ شوال سنه ۱۰۶۰ هجری.


It has been finished: the comparison of twelve teachings from the book Dabestān, composed by the elder of the truth-seekers, the leader of the scrutinizers, perfect mystic, the arrived sufi, the sage of the house of wisdom, where to perceive the truth, the recognizer of the teachings of the honored Absolute Existent, confirmed by praised affirmations, the arch-healer, the master Mirzā Ẕolfaqār Āẕar-sāsānī, with the pen name Mūbed, may God elongate his age, (which) was authored in the year 1060. I corrected it to the limit of my endurance, and I did (this) as much as constraints allowed, and to the extent of my recognition. I noted (the differences) at the margin with the character mīm. Hopefully, it will stay in God’s safety, away from error. If God wills, may what will be authored later be recorded. The humble student, Mağd-al-dīn Muḥammad, is the one who compared this magnum opus of the honored master. Redacted on 23 Šawwāl 1060 h. [October 19, 1650].57


This epilogue provides a definite answer to the question of the text’s authorship. The author was a certain Mīrzā Ẕolfaqār Āẕar Sāsānī, who wrote under the pen name Mūbed.58 It moreover gives a terminus ante quem for authoring the Dabestān-e maẕāheb as well as for the other Āẕar Kaivānī treatises mentioned in this book. Hence, the Dabestān-e must have been authored before 1060/1650. A terminus post quem of 1653 for the Dabestān-e maẕāheb has been already inferred from the events mentioned in its edition text (Keyḫosro 1362, 1/122, 2/20): Welcome to the paradox! The inconsistency consists in major differences between the text of this manuscript (Āẓar Sāsānī 2010) and the published text of the Dabestān (Keyḫosro 1362, 1362). Comparing the volume of Reżāzāde Malek’s edition with this manuscript shows that the text was expanded by ca. 16.4%, or about 23,000 tokens.59


In his notes to the edition of the Dabestān-e maẕāheb, Reżāzāde Malek lists the dates explicitly mentioned in the Dabestān-e maẕāheb (Keyḫosro 1362, 2/10-16). To find the terminus post quem for the Dabestān-e maẕāheb I went through this list in reverse chronological order and checked for the existence of the passages involving these dates in the manuscript from 1060/1650. The passages consisting of the dates 1063/1653 and 1061/1651, which are attested in the edition, are not present in this manuscript.60 The migration of Šāh-Badaḫšī to India, his initiation into the Mīr-Qāderī order and his acceptance of Moḥyī-al-dīn Moḥammad as a student, which is the last event in Reżāzāde Malek’s list, are absent in the manuscript as well.61 By this, the latest date mentioned in the manuscript is 1059/1649. The corresponding passage reads:


پیکرپژوه و جهان‌نورد دو تن بودند از پیکری‌کیشان که در جدول‌کشی و تصویر و نقاشی بی‌بدل بودند. نامه‌نگار بسال پنجاه و نهم در گجرات من اعمال پنجاب هر دو را یافت.


Peykarpažūh and Ğahān-navard were two persons of the group of Peykarī, who were unique in creating rule-borders, illustrating and painting. This author visited both of them in the 59th year [= 1649 M.] in Gujarat from Punjab.62


Two other passages in the book give information about the period of its writing:


و اکنون که سنه هزار پنجاه و پنج هجریست پسر مهروان که اوبرجی جانشین اوست […]


And now, the year 1055 Hiğrī [= 1645 M.], the son of Mihravān, whom Ūbarğī (?) succeeds, […]63


اکنون که هنگام نبشتن این نامه است و سال هجری بهزار و پنجاه و پنج رسیده‌ […]


And now that the time of written of this book, the Hiğrī year 1055 (1645 m.) has come […]64


At the beginning of the second chapter of the book in its published edition, which is about Hindus, the author adds an editorial note revealing that the author visited a group of Hindus in 1063/1653. This visit led to revision of this chapter of the book specifically. The author writes at the end of this editorial note: “Consequently a difference occurred between the first and second edition [lit. order].”65 Consequently, the manuscript of 1060/1650 should represent the manuscript of the first recension of the book, while later manuscripts represent the latter recension after the year 1063/1653. The author must have worked on the text of the first recension for a period of at least five years, from 1055/1645 to 1060/1650. The differences between the two recensions of the text are not limited to the chapter on Hindus, although this chapter remains the most heavily revised part of the book. The author enlarged this chapter in his second recension by about 10,000 tokens. This means that he added another 13,000 tokens to other parts of his book in its second recension.


The epithet āẕar in the name of the probable founder of the school, Āẕar Kaivān, helps to illuminate the interreligious contact between the school with Zoroastrianism. According to the Dabestān-e maẕāheb the epithet āẕar, ‘fire,’ was assigned to the names of all of his precedents as well. Moreover, the author of the Dabestān-e maẕāheb, another prominent member of the school, also bore the title āẕar. One of the names given by the Dabestān-e maẕāheb to the school, Āẕarīyān, seems to be connected to this epithet. The epithet in the name of some members of the school, and the importance of fire in religious theories of the school, is emphasized in Āẕar Kaivān’s genealogy as well as in the name Āẕarīyān for the school.


On his expedition to Gujarat, Akbar made the acquaintance of Mūbed Meherjī Rānā’ and invited him to the courtly discussions of 1578 and 1579. Consequently, he spent 1578–79 in Fathpur as the first representative of a non-Islamic religion in order to participate in the discussions in the ʿebādat ḫāna ‘House of Worship’ founded by Akbar. In 1581–82, Akbar introduced a form of the Zoroastrian cult of fire to his court. The sojourn of Meherji Rānā at the court was presumably influential in this measure.66 Afterwards, the compatibility of this cult of fire with Islamic monotheism was intensively discussed at the court. The Zoroastrian theological interpretation of fire as the everlasting symbol of God on earth must have ensured that it took a prominent place in the theological discourse of this period. Consequently, the bearers of the epithet āẕar were connected to ancient Iranian cultural assets, as well as endowed with theological prestige. Therefore, I would like to propose the date of Akbar’s introduction of the cult of fire at his court as the terminus post quem for the authoring of the Dasātīr. Accordingly, it can be hypothesized that the Dasātīr was written after 1581-82. Because of the influence of Sanskrit on the heavenly language of the Dasātīr (Mojtabaʾī 1994), we can assume that it was authored after the migration of Āẕar Kaivān to Patna, assuming Āẕar Kaivān was its author. By assuming that Āẕar Kaivān migrated to Patna in 1001/1593 we can even limit the terminus post quem to this date. We can regard the date of the first recension of Dabestān-e maẕāheb, 1060/1650, or even the date of death of Āẕar Kaivān, 1028/1618, as the terminus ante quem of the Dasātīr. Subsequently, the Dasātīr must have been authored between 990/1581-2 and 1060/1650, or Āẕar Kaivān must have authored it between 1001/1593 and 1028/1618. The encounter of the Dasātīr with Zoroastrian Middle Persian literature, thus, must have occurred in the same period, and likely took place in Patna in India.


Were the Āẕar Kaivānīs the first non-Zoroastrian New Persian speakers who detected Middle Persian texts and developed a fascination for it? This was the assumption in the scholarship of the last centuries. Recently, Ali Ashraf Sadeghi (2020) made a significant discovery which sheds light on the acquaintance of early modern New Persian-speaking literates with Middle Persian literature. Previously, the scholarship assumed that the Borhān-e Qāṭeʿ was the oldest dictionary citing ‘dasātīrī’ terms. Sadeghi shows that the ‘dasātīrī’ terms are actually older than the Dasātīr. According to him, the Farhang-e Moʾaiyad al-fożalāʾ, authored by Moḥammad b. Lād Dehlavī in 925/1519, had already cited such words at least 65 years before the Dasātīr saw the light of day. Sadeghi shows, moreover, that the Farhang-e Moʾaiyad al-fożalāʾ cites not only ‘dasātīrī,’ i.e., artificially antiquated New Persian words famously used in the Dasātīr, but also Middle Persian lexemes. He lists, for example, odardan ‘to pass away’ (gained from MP widardan <wtltn>)67, basrīyā ‘meat’ (gained from MP gōšt <BSLYA>), baytā ‘house’ (gained from MP xānag <BYTA>), pāteprās ‘punishment’ (gained from MP pādifrāh <pʾtplʾs>), čīčast ‘mountain’ (gained from MP čēčast <čyčst> ‘a mythical sea’), and finally čīnvad ‘bridge to the hereafter’ (gained from MP činwad (puhl) [cynwt] ‘bridge to the hereafter’). This evidence asserts that the New Persian speaking literates in India were already acquainted with and fascinated by Middle Persian in the first decades of the sixteenth century. The Āẕar Kaivānīs were thus not the initiators of this contact with Zoroastrianism and the Zoroastrian Middle Persian—they were its consumers. As early as 925/1519, there was contact between Muslim literates and Zoroastrian texts in India. The Āẕar Kaivānīs, however, extended this literary contact to a religious one.

Conclusions: the Dasātīr and Secrecy


As we saw above, the Avestan texts are represented in younger Zoroastrianism as concealed texts, and Avestan as a celestial language which was spoken only in the communication of Ahura Mazdā and Zarathustra. This perspective, however, was not adopted by older Zoroastrianism when Avestan was still used for text production. Even in the Sasanian and early Islamic periods, the Avestan language was not perceived or represented as a secret language. The Zoroastrian priests were engaged in the translation of, and commentary on, these texts. Because of the reduced competence of the priests in understanding the Avestan language in the first half of the second millennium A.D., perspectives on the Avestan language underwent significant change. Avestan texts came to be perceived as secret texts which were not supposed to be understood by Zoroastrians, and which were accessible only through translations and commentaries. In this way, the Zoroastrians in this period constructed an ‘other-world’ by relocating the Avestan language to the transcending divine sphere. They did not use this emerging secrecy to establish an insider-outsider distinction. Rather, they highlighted the inherent potential of a secret language for communication with the divine sphere, modeled upon Zarathustra’s communication with Ahura Mazdā and unceasingly re-exemplified in Zoroastrian rituals, i.e., in priests’ communication with the divine world.


By adopting the concept of a secret, celestial language from Zoroastrian Zand literature, the Āẕar Kaivānīs remained within the Zoroastrian conceptual framework of secrecy. The Āẕar Kaivānīs did not use the secret language to establish an in-group / out-group distinction vis-à-vis other religions, because they did not claim the ability to understand and translate it. Interestingly, they also made clear that the competence to understand and translate the heavenly language was restricted to older prophets; not even Āẕar Kaivān or the author of the Dasātīr claimed this competence for himself. The Āẕar Kaivānīs even dispensed with claims of access to the heavenly language, which in Zoroastrianism was an intra-religious demarcation parameter between a group of specialists and other Zoroastrians. It is true that, in the early modern period, they did not know that the Zoroastrian priests were able to translate and comment the Avestan texts in the Sasanian period. Nevertheless, they hypothetically could have constructed their Dasātīr in such a way as to show that a specific group of their circle would have access to the language of heaven. Hence, we can conclude that the Āẕar Kaivānīs did not use the secrecy of their celestial language for purposes of inter- or intra-religious demarcation or to gain intra-religious authority or inter-religious superiority.


Rather, the Āẕar Kaivānīs’ strategy of secrecy seems to be a sort of double coding (Boneberg 2005, 461). Knowledge is encoded on two layers: communicated in translation and commentary as well as encoded in celestial language. The Āẕar Kaivānīs developed a strategy of secrecy rather than distinction. They used secrecy to construct an other-world which cannot be reached directly, but only through the mediation of translation and commentary. This secrecy is not characterized as a mode of exclusion; in contrast, it is extremely inclusive. The constructed other-world applies to all religious traditions in the same way and is or is not available to them to the same degree. Their secrecy is not a concealment of knowledge but a sharing of the concealed. Dasātīr’s approach to secrecy is in perfect accord with the religious discourse emerging at the court of Akbar, namely dīn-e elāhī.


This investigation shows that the contact with the Zoroastrian Middle Persian texts was established in the early Modern Persian speaking elite circles and outside of the religious field. Presumably, it was the lexicographical interest which first led to the re-discovery of Middle Persian as an antique form of New Persian. To include noble forgotten Persian words in their dictionaries, the lexicographers gained Middle Persian lexemes from the Zoroastrian texts. The Āẕar Kaivānīs presumably became acquainted with the Middle Persian literature through these lexicographical activities in India. They, however, extended this language contact to a religious contact. They created a heavenly language and a heavenly book after the Zoroastrian Zand texts. They avoided Arabic words and created a form of Persian imitating Sasanian Middle Persian. Whereas the form of Zoroastrian literature must have strongly influenced Āẕar Kaivānī literature, their contents do not seem to have been influential for this school.



ʿĀbedī, Seyyed Amīr Ḥasan. 1383. “Mīr Ẓolfaqār Āẕar Sāsānī: Moʾallef-e Dabestān-e maẕāheb.” Translated by Seyyed Ḥasan ʿAbbās. Nāme-Ye Anğoman 4: 161–64.
Abolqāsemī, Moḥsen. 1386. Dīn-hā va kīš-hāye īrānī dar dawrān-e bāstān be ravāyat-e Šahrastānī. Tehran: Sepīdrūd.
Andrés-Toledo, Miguel Ángel. 2012. “The Avestan Manuscripts T55a_681 (Sanskrit Yasna) and T55b_613 (Pahlavi Yasna) of the First Dastur Meherji-Rana Library of Navsarī.” Avestan Digital Archive Series 51. http://www.avesta-archive.com.
———. 2015. “Primary Sources. Avestan and Pahlavi.” In The Wiley Blackwell Companion to Zoroastrianism, edited by Michael Stausberg and Yuhan Sohrab-Dinshaw Vevaina, 519–28. Hoboken: John Wiley & Sons Inc.
Anquetil-Duperron, Abraham Hyacinthe. 1771. Zend-Avesta: Ouvrage de Zoroastre. 2 vols. Paris: N.M. Tilliard.
Aoki, Takeshi. 2000. “The Role of Āzar Kaivān in Zoroastrian and Islamic Mysticism.” In K. R. Cama Oriental Institute. Third International Congress Proceedings, 259–77. Mumbai: K. R. Cama Oriental Institute.
———. 2016. “A Zoroastrian Refutation of the Muʿtazilite Theology, with an Edition of ʿUlamā-Ye Islām.” Journal of Central Eurasian Studies 4: 1–27.
Āẓar Sāsānī, Mirza Zulfaqar. 2010. Dabestān-e Mazāheb. Facsimile of Manuscript 1060 A.H. / 1650 A. D. Edited by Karim Najafi Barzegar. New Dehli: Iran Culture House.
Baġdādī, ʾAbū Manṣūr ʿAbd-al-Qāhir b. Ṭāhir b. Muḥammad al-. (1328) 1910. al-Farq bayn al-firaq wa-bayān al-firqa al-nāğīya minhum. Edited by Muḥammad Badr. Qāhira: al-Maʿārif.
Bartholomae, Christian. 1915. Die Zendhandschriften der K. Hof- und Staatsbibliothek in München. München: Palm in Komm.
Bayevsky, Solomon. 1999. “Farhang-e Jahāngīrī.” EIr. http://www.iranicaonline.org/articles/farhang-e-jahangiri.
Boneberg, Hemma. 2005. “Geheimhaltung.” In Metzler Lexikon Religion: Gegenwart – Alltag – Medien, edited by Christoph Auffarth, Jutta Bernard, and Hubert Mohr, 1:460–62. Stuttgart: Metzler.
Cantera, Alberto. 2004. Studien Zur Pahlavi-Übersetzung Des Avesta. Iranica 7. Wiesbaden: Harassowitz.
Dehghani Farsani, Yoones, and Kianoosh Rezania. 2020. “Ibn Al-Malāḥimī on Zoroastrianism.” Iranian Studies 53: 1–37. https://doi.org/10.1080/00210862.2020.1713058.
Dehlavī, Moḥammad b Lād. n.d. “Farhang-e Moʾaiyad al-fożalāʾ.” Edited by Ervad Bamanaji Nasarvanji Dhabhar. Library, Museum and Document Center of Iran Parliament, Tehran (classification number 15019).
Dhabhar, Ervad Bamanaji Nasarvanji, ed. 1909. Saddar Nasṛ and Saddar Bundehesh. Persian Texts Relating to Zoroastrianism. Bombay: British India Press.
Douglas, Mary. 2007. Thinking in Circles: An Essay on Ring Composition. Terry Lecture Series. New Haven: Yale University Press.
Enğū Šīrāzī, Mīr Ğamāl-al-dīn Ḥosayn b. Faḫr-al-dīn Ḥasan. (1351) 1972. Farhang-e Ğahāngīrī. Edited by Raḥīm ʿAfīfī. Mašhad: Danešgah-e Mašhad.
Ernst, Carl W. 2017. “Concepts of Religion in the Dabistan.” In It’s Not Just Academic! Essays on Sufism and Islamic Studies, 437–62. Thousand Oaks, California: SAGE Publications Inc.
Ferdowsi, Abu’l-Qasem. 1988–2008. The Shahnameh (Book of Kings). Edited by Djalal Khaleghi-Motlagh. 8 vols. Persian text series 1. New York: Bibliotheca Persica.
Galewicz, Cezary. 2009. A Commentary in Service of the Empire. Sāyaṇa and the Royal Project of Commenting on the Whole of the Veda. Publications of the de Nobili Research Library 35. Wien: Sammlung de Nobili.
Grobbel, Gerald. 2007. “Das Dabistān-i Maḏāhib und seine Darstellung der Religionsgespräche an Akbars Hof.” In Islamische Grenzen Und Grenzübergänge, edited by Benedikt Reinert and Johannes Thomann, 85–130. Schweizer Asiatische Studien. Monographien; Etudes Asiatiques Suisses. Monographies 44. Bern/New York: Peter Lang.
Hottinger, Arnold. 1998. Akbar der Grosse, Herrscher über Indien durch Versöhnung der Religionen. München: Wilhelm Fink.
Ḫwārazmī, Maḥmūd b. Muḥammad al-Malāḥimī, and Wilferd Madelung. 2012. Kitāb al-muʻtamad fī uṣūl al-dīn. Classical Muslim Heritage Series 1. Tehran: Mīras̲-e Maktoob.
Isfarāʾinī, ʾAbī al-Muẓaffar al-. 1374. al-Tabṣīr fī l-dīn wa-tamyīz al-firqat al-nājīya ʿan al-firaq al-hālikīn. Edited by Muḥammad Zāhid b. Ḥasan al-Kawtharī. Qāhira: al-Khānjī.
Jones, William. 2013. The Works of Sir William Jones: With the Life of the Author by Lord Teignmouth. Edited by Lord Teignmouth. Cambridge: Cambridge University Press.
Keyḫosro, Esfandīyār. 1362. Dabestān-e maẕāheb. Edited by Raḥīm Reżāzāde Malek and Wilferd Madelung. 2 vols. Adabīyāt-e ʾAsāṭīṛī 1.2. Tehran: Ṭahūrī.
Modak, B.R. 1992. “Sayana.” In Encyclopaedia of Indian Literature, edited by Mohan Lal, V:3885–86. Delhi: Sahitya Akademi.
———. 1995. Sayana. Makers of Indian Literature. New Delhi: Sahitya Akademi.
Modi, Jivanji Jamshedji. 1903. The Parsees at the Court of Akbar and Dastur Meherjee Rânâ: Two Papers Read by Jivanji Jamshedji Modi, Before the Bombay Branch Royal Asiatic Society on 19th December 1901 and 13th July 1903. Bombay: Bombay Branch Royal Asiatic Society.
Mojtabaʾī, Fatḥ-allāh. 1994. “Dasātīr.” EIr 7: 84.
Morgenstierne, Georg. 1926. Report on a Linguistic Mission to Afghanistan. Instituttet for Sammenlignende Kulturforskning. Serie C 2-1. Oslo: Aschehoug.
Mulla Firuz b., Kaus. 1818. The Desatir or Sacred Writings of the Ancient Persian Prophets; in the Original Tongue; Together with the Ancient Persian Version and Commentary of the Fifth Sasan. Bombay: Courier.
Müller, Max. 1849. Rig-Veda-Sanhita, the Sacred Hymns of the Brahmanas: Together with the Commentary of Sayanacharya. 6 vols. London: W. H. Allen and Co.
Rezania, Kianoosh. 2010. Die zoroastrische Zeitvorstellung: eine Untersuchung über Zeit- und Ewigkeitskonzepte und die Frage des Zurvanismus. GOF III / NF 2 7. Wiesbaden: Harrassowitz.
Sadeghi, Ali Ashraf. 2020. “Are All the Dasātiri Words Attributabler to the Āzarkeyvān Religious Group?” Journal of Iranian Studies (Iran Kenkyu) 16: 96–100.
Šahrestānī, Muḥammad b. ʿAbd al-Karīm. (1395) 2016. Tarğoma-ye ketāb-e al-Milal wa-l-niḥal: nosḫa-ye bargardān-e dastnevīs-e šomāra-ye 2371 Ketābḫāna-ye Ayāṣūfiyā (Estānbūl). 1st ed. Vol. 290. Mīrās̱-e Maktūb; Mağmūʻa-ye nosḫa-bargardān, 290. 24. Tehran: Mīrās̱-e Maktūb.
Šahrestānī, Muḥammad b. ʿAbd al-Karīm. 1961. al-Milal wa-l-niḥal. Edited by Muḥammad Kīlānī. Vol. 2. Beirut: Dār al-Ma‛rifa.
Shaked, Shaul. 1994. “Some Islamic Reports Concerning Zoroastrianism.” JSAI 17: 43–84.
Shea, David, and Anthony Troyer. 1843. The Dabistan or School of Manners, Translated from the Original Persian, with Notes and Illustrations. Translated by from. 2 vols. Paris: Duprat.
Sheffield, Daniel J. 2014. “The Language of Heaven in Safavid Iran: Speech and Cosmology in the Thought of Āẕar Kaivān and His Followers.” In No Tapping around Philology. A Festschrift in Honor of Wheeler McIntosh Thackston Jr.’s 70th Birthday, edited by Alireza Korangy and Daniel J. Sheffield, 161–83. Wiesbaden: Harrassowitz.
———. 2018. “Lord of the Planetary Court: Revising a ‘Nativist Prophet’ of Early Modern Iran.” In Studying the Near and Middle East at the Institute for Advanced Study, Princeton, 1935-2018, edited by Sabine Schmidtke, 455–66. Piscataway, NJ: Gorgias Press.
Unvâlâ, Ervad Manockji Rustamji. 1922. Dârâb Hormazyâr’s Rivâyat. 2 vols. Bombay.
Zadeh, Travis. 2012. The Vernacular Qurʾan. Translation and the Rise of Persian Exegesis. Qurʾanic Studies Series 7. Oxford: Oxford University Press.

  1. From the contents of the Dabestān-e maẕāheb, Carl Ernst (2017, 440) concludes that the title of the book can alternatively be translated as The School of Theologies.↩︎

  2. Takeshi Aoki (2000, 263) dates Āẕar Kaivān’s migration to India in the period between 1573 and 1580.↩︎

  3. Three names Āzādān, Sorūšān, Hūšīyān are absent in the edition of Keyḫosro (1362), 5f. I quote the Dabestān-e maẕāheb after the facsimile publication of its oldest manuscript (Āẓar Sāsānī 2010) as well as its edition (Keyḫosro 1362). An English translation of the book can be found in Shea and Troyer (1843).↩︎

  4. For a detailed survey on the concept of religion in the Dabestān-e maẕāheb, see Ernst (2017, 438–46).↩︎

  5. Sir William Jones, the British orientalist, was the first to draw attention to this book and consequently to Āẕar Kaivān and this school by praising the Dasātīr in 1789 (Jones 2013).↩︎

  6. Āẕar-sāsānī (2010, 8r); parallel to Keyḫosro (1362, 10). Depending on how the verbs are to read: bowad and na-mī-mānad or būd and na-mī-mānd.↩︎

  7. All translations into English are by the author unless indicated otherwise.↩︎

  8. In his edition of the text, Mulla Firuz uses two signs to mark the division between the phrases in the celestial language on the one hand and their translations and commentaries on the other. A similar representation can be found in some lithographic reprints of the book, which I found in the Library, Museum and Document Center of Iran Parliament, Tehran (classification number 2937 and 128162). In the book with the classification number 86831 from the same library, moreover, the text in the celestial language has been partly written on the margin. In this book and in the one with the classification number F7474, the word bayān or šarḥ separates the translation from the commentary. In number F4609, the text in celestial language is written in red.↩︎

  9. Aoki (2000, 264f.) suggests that the Āẕar Kaivānīs used Arabic words in their works before their emigration to India. According to him, their reservation against the use of Arabic words first arose in India.↩︎

  10. Āẕar-sāsānī (2010, 104v, ll. 9–15); parallel to Keyḫosro (1362, 113).↩︎

  11. For an overview to Sāyana’s life and works see Modak (1992, 3885–86.) and Modak (1995).↩︎

  12. In the exegetical works assigned to him, his brother, Mādhava, as well as more assistants seem to have been involved. For an elaborated investigation of his commentary project, see Galewicz (2009).↩︎

  13. For an edition of Ṛgvedasaṃhitābhāṣya, see Müller (1849).↩︎

  14. See Galewicz (2009, 295) and figure 1.↩︎

  15. This structure can be called ring composition; for this, see the classic work of Mary Douglas (2007).↩︎

  16. The representation of Ṛgvedasaṃhitābhāṣya’s structure should, moreover, demonstrate that the linear sequence of original and commentary is not the only possible form for exegetical literature, even if it is the simplest and most manifest.↩︎

  17. Zadeh (2012, 266); for some examples of manuscripts, see Zadeh (2012, figs. 2, 10).↩︎

  18. See Anquetil Duperron’s (1771, Ouvrage de Zoroastre, 2:1.1/iii) hint regarding the language of Zend-Avesta as an old language of north Persia, as well as Morgenstierne’s (1926, 29–30) contextualization of Avestan in east Iranian languages.↩︎

  19. We can find the same opinion on Avesta in the older Zoroastrian literature. Identifying a source that is chronologically close to the Dasātīr demonstrates that the Āẕar Kaivānīs may have received this opinion from Zoroastrian New Persian literature.↩︎

  20. UIbdR, 85; in the original bidānand instread of bidānad.↩︎

  21. For an exhaustive study on the Pahlavi translation of the Avesta, see (Cantera 2004).↩︎

  22. The term zand, moreover, designates the texts based on the Pahlavi translation of the Avesta. This part of Zand literature, however, is not decisive for our discussion here.↩︎

  23. The meaning of the text is not important for our discussion.↩︎

  24. The text is transcribed after the ms. T55 (Andrés-Toledo 2012). One folio of this manuscript can be seen in Figure 2.↩︎

  25. Dhabhar (1909, 66); in the original vājit instead of vājib.↩︎

  26. Āẕar-sāsānī (2010, 90v), parallel to Keyḫosro (1362, 111).↩︎

  27. This is the case in all Zoroastrian manuscripts of the Pahlavi translation; I did not have the chance to check all manuscripts of the Dasātīr. In the case of the Dasātīr, however, I do not see a necessity for such a double check because these three components undoubtedly belong together on the conceptual level. If one assumes that the New Persian text constitutes the starting point of the Dasātīr, it must remain bound to its conversion into the constructed language. From this perspective it is impossible to present these three components separately in the construction of the Dasātīr.↩︎

  28. On this dictionary, see Bayevsky (1999).↩︎

  29. Modi (1903, 92–93) uses the attestation of a Persian Revāyat, a correspondence between Irani and Parsi Zoroastrian priests, to show that Ardašīr left India in 1597. Therefore, he must have been located, for an unknown period of time until 1597, at Akbar’s court.↩︎

  30. On the significance of this latter dictionary see below.↩︎

  31. Dehlavi (n.d., 436). This dictionary defines pāzand similar to Zandavestā.↩︎

  32. Keyḫosro (1362, 111–12); this passage is absent in the first recension of the work (Āẓar Sāsānī 2010).↩︎

  33. See e.g. Grobbel (2007, 99); Sheffield (2018, 457–58).↩︎

  34. Āẕar-sāsānī (2010, 57v–95v) = Keyḫosro (1362, 72–118).↩︎

  35. Āẕar-sāsānī (2010, 58r–74v) = Keyḫosro (1362, 72–93).↩︎

  36. Āẕar-sāsānī (2010, 75v–81r) = Keyḫosro (1362, 94–100).↩︎

  37. Āẕar-sāsānī (2010, 82r–90v) = Keyḫosro (1362, 101–11).↩︎

  38. Keyḫosro (1362, 87); this passage is absent in the first recension; see fol. 72r in (Āẓar Sāsānī 2010).↩︎

  39. Āẕar-sāsānī (2010, 90v, l. 20–91r, l. 15), parallel to Keyḫosro (1362, 112–13).↩︎

  40. Sheffield (2018, 458); Āẕar-sāsānī (2010, 31r–31v) = Keyḫosro (1362, 36).↩︎

  41. For the case of Arabic texts, and al-Šahrestānī’s heresiography in particular, see below in this section.↩︎

  42. The term damān-keš occurs in the Dasātīr only in the phrase bīmāye va damān-keš attributing creator (D., 78, 130, 135). We can derive the meaning of these adjectives from the following phrase, D., 149: به یک تاب خدا دو جهان اشکارا شد که یکی جهان بیمایه و هنگام و دوم گیتی مایه دار باشد هر دورا هستی از پرتو خورشید گوهر دادار است. / “A radiance of God originated both worlds. One is the immaterial [bīmāye] world without duration [()-hengām], the second one is material universe. Both have their existence from a beam of the sun of creator’s essence.↩︎

  43. D, 52, translated by Mulla Firuz.↩︎

  44. All three phonemes are represented with the same letter in the Pahlavi script.↩︎

  45. On this, see this section below.↩︎

  46. For the history of research on the Zurwān myth in the Iranian Studies, which started two centuries later, see Rezania (2010, 12–43); an interpretation of the myth can be read in Rezania (2010, 169–200).↩︎

  47. UI, §§21f. = Unvâlâ (1922, 2/75, ll.17–19, 76, 1–4).↩︎

  48. See manuscript M55, edited by Bartholomae (1915, 113–14).↩︎

  49. As an example, I can mention the sections about the belief system of the Mazdakites. The text in the Dābestān-e maẕāheb (Āẓar Sāsānī 2010, 97r; Keyḫosro 1362, 119) strongly resembles the corresponding passages from al-Šahrestānī’s Arabic text (Abolqāsemī 1386, 153–54; Muḥammad b. ʿAbd al-Karīm. Šahrestānī 1961; Shaked 1994). The New Persian translation of this Arabic book from the sixth century H. (Muḥammad b. ʿAbd al-Karīm Šahrestānī [1395] 2016, vol. 290, fol. 117r and v), however, differs in some places from both these texts, e.g., in the number of spiritual managers, 13 in contrast to 12, and their order. As broadly discussed, al-Šahrestānī lists here 13 elements but gives their number as 12; the Persian translation corrects their number to 13. It nevertheless enumerates 14 elements because davande is repeated twice by mistake; for another citation from al-Šahrestānī in the Dabestān-e maẕāheb, see Ernst (2017, 443–44).↩︎

  50. See passages 14, 20-22 (Abolqāsemī 1386, 135–36).↩︎

  51. These include al-Isfarāʾinī (1374, 132), al-Baġdādī ([1328] 1910, 347), and even the exhaustive theological discussion of al-Malāḥimī al-Ḫwārazmī (2012, 638ff.). On this, see Dehghani Farsani and Rezania (2020).↩︎

  52. Āẕar-sāsānī (2010, 81v, l. 15–82r, l. 1); parallel to Keyḫosro (1362, 100–111). This passage alludes to the Zoroastrian narration recounted by Ferdowsi (1988–2008, 5/81-4). According to the narration of the ‘Cypress of Kašmar,’ Zarathustra brought a sapling of a noble cypress (sarv-e āzāda) from paradise and gave it to Goštāsp, who planted it in front of the first fire temple in Kašmar in Khorasan. In only a few years, it grew into a huge, beautiful cypress, serving as a focal point for pilgrimage. The sources of the Islamic period, e.g., Ṯ̄aʿālibī, report that the caliph al-Mutawakkil wished to see this cypress. As it was not possible for him to travel to Nishapur, he commanded his governor in Khorasan to cut the tree and to send it to Baghdad. The Zoroastrians tried to prevent the inauspicious felling of their cypress by offering the caliph 50,000 dinars, which he rejected. 1300 camels carried the pieces of the cypress to the caliph, who was assassinated just one day before the convoy arrived in his capital; see Aʿlam 1993.↩︎

  53. Āẕar-sāsānī (2010, 91r, l. 15 – 91v, l. 1); parallel to Keyḫosro (1362, 113).↩︎

  54. Keyḫosro (1362, 101); this passage is absent in Āẕar-sāsānī (2010).↩︎

  55. Āẕar-sāsānī (2010, 6v) = Keyḫosro (1362, 8); the first smallest units, day, month and year, are not mentioned in Āẕar-sāsānī (2010).↩︎

  56. Āẕar-sāsānī (2010, 302); see ʿĀbedī (1383, 162) as well.↩︎

  57. Āẕar-sāsānī (2010, 302); see ʿĀbedī (1383, 162) as well.↩︎

  58. Āẕar-sāsānī (2010, 13–15).↩︎

  59. I estimate the number of tokens in the manuscript as approx. 140,000, in the edition around 163,000. The estimation for the first text is based on the count of words of its first 50 folios; for the second text, it relies on the word count of a digital version of the text.↩︎

  60. The first date is attested in Keyḫosro (1362, 122, ll. 3–8) and is expected on Āẕar-sāsānī (2010, fol. 99r); the second date is attested in Keyḫosro (1362, 18–19, ll. 27–4) and expected on Āẕar-sāsānī (2010, fol. 16r).↩︎

  61. It is attested in Keyḫosro (1362, 359, ll. 11–19) and expected on Āẕar-sāsānī (2010, fol. 295v).↩︎

  62. Āẕar-sāsānī (2010, fol. 55v, ll. 8–11), Keyḫosro (1362, 69, ll.9–11).↩︎

  63. Āẕar-sāsānī (2010, fol. 142r, ll. 8–9), Keyḫosro (1362, 207, l.11).↩︎

  64. Āẕar-sāsānī (2010, fol. 106r, ll. 12–14), Keyḫosro (1362, 135, ll.7–8).↩︎

  65. Keyḫosro (1362, 1/122, ll. 7–8): “لاجرم میان ترتیب اول و ثانی مباینتی روی داد.↩︎

  66. See Modi (1903, esp. 152-58); Hottinger (1998, 116–17, 129–30).↩︎

  67. We should take into consideration that the Pahlavi script often uses the character <l> to represent the phoneme r.↩︎