On the question of domestic slaves in late medieval and early modern Zoroastrianism

Abstract

This article collects and analyses passages about male and female domestic slaves in the Persian Rivāyats. The Rivāyats consist of correspondence between Iranian and Indian Zoroastrians (Parsis) from the late fifteenth to the eighteenth centuries ce. In these letters, Parsis sought the opinions of Iranian Zoroastrians on various doctrinal and ritual issues. The passages in question cover a range of subjects, including the issue of converting household slaves to Zoroastrianism, their participation in domestic religious ceremonies, the exposure of their dead bodies in the towers of silence, and marrying female slaves. These references to slaves challenge the conventional narrative that pre-modern Zoroastrians were oppressed, marginalized, and poor communities. This narrative has overshadowed these pieces of evidence and has caused them not to be studied seriously. This paper seeks to go beyond this traditional reconstruction by examining these texts based on their context. The passages reflect the actual socio-religious issues of Zoroastrians, especially Parsis, and demonstrate their participation in the slave-owning milieu of late medieval and early modern Gujarat and Iran rather than mere anachronistic elements or rhetorical tools reflecting a scholastic treatment of a defunct legal question.

When it was time to start saying prayers dustoorji told me to go inside. Later, bai told me that was because Parsi prayers are so powerful, only a Parsi can listen to them. Everyone else can be badly damaged inside their soul if they listen.

Rohinton Mistry, Tales from Firozshah Baag, 63

Introduction

Rohinton Mistry, an acclaimed Parsi novelist, writes about the life of Jacqueline, a Catholic ayah (i.e. nursemaid or female servant) from Goa, in one of the stories from his Tales from Firozshah Baag (originally published in 1987). The collection revolves around Parsi life in a housing colony (baug) in Bombay. Jacqueline, living in a Parsi household, tells her life story from her own perspective. Although her real name is Jacqueline, she is mistakenly called Jaakaylee by everyone in the Parsi compound. One Christmas Day, when she encounters a ghost on the staircase, the people in the baug label her as crazy and subject her to scornful mockery. A year later, in an unexpected turn of events, the mistress of the house (bai) and her husband attend a New Year's Eve dance. Upon reaching home, the bai, catching sight of Jacqueline standing on the balcony lost in her childhood memories, covered by a white bedsheet, mistakes her for a ghost. This encounter of bai with a “ghost” creates a bond between her and Jacqueline, blurring the boundaries of their respective roles as mistress and servant. To rid the balcony of its “ghost”, a neighbouring priest suggests that the bai perform a Parsi prayer ceremony (jashan) there. The opening epigraph of this article hints at Jacqueline's exclusion from this ceremony. In this captivating tale, Mistry skilfully weaves the lives of non-Parsi servants, often overlooked in Parsi households, into the fabric of daily life among modern Parsis in Bombay.¹

The focus of this paper is on the predecessors of Jacqueline – domestic slave-servants who worked and lived in Zoroastrian houses during the late medieval and early modern periods. It explores their complex socio-religious status, as reflected in a few questions and answers found in the Persian Rivāyats. While Jacqueline's story is narrated from her own perspective, it is important to recognize that it is ultimately a construction by Mistry. Similarly, in the sporadic references to domestic slaves in the Rivāyats, they are only presented through the lens of their Zoroastrian masters.

The use of the term “slave” throughout this article may be perceived as inappropriate by certain readers, particularly those within contemporary Zoroastrian communities in Iran, India, and the diaspora. It could evoke associations with the American form of slavery, which has significantly influenced the definition of slavery even within academic circles.² This definition has limited the term's meaning to a single cultural context. My assumption is that slavery encompassed a broader range of cultural manifestations that can and should be compared to one another. There were indeed both similarities and differences between various models of slavery.³ It is essential to clarify that when the present author refers to slaves in the Iranian or South Asian context, it does not necessarily equate to the experiences associated with black slaves under the American model.

Despite the recent surge in interest among students of Iranian studies, a comprehensive history of slavery in the Iranian world has yet to be written.⁴ What is certain is that pre-modern Iranian and Indian societies were both slave-holding societies for an extended period of time.⁵ The legal trade in slaves and their ownership persisted in India until the nineteenth-century abolitionist movements.⁶ In Iran, the process leading to the abolition of slavery began with a royal decree from Muhammad Shah Qajar (r. 1834–48) in 1848, which prohibited the slave trade in the Persian Gulf. However, this farmān had a limited scope and was primarily a response to the pressures imposed by Western powers, particularly Britain, on the Qajar court.⁷ The situation gradually changed during the second half of the nineteenth century, eventually culminating in the complete legal suppression of the institution in the early twentieth century.⁸

Some studies have been conducted on the institution of slavery and its related terminologies in pre-Islamic Iran within the wider context of research on ancient Iranian society.⁹ The primary textual source for this subject has been Mādayān ī Hazār Dādestān (Book of a Thousand Judgments, hereafter MHD), a Zoroastrian legal compendium written in Middle Persian (Pahlavi), which contains numerous regulations concerning slaves. This text is commonly believed to have been composed during a period when Zoroastrian political, legal, and social dominance was at its peak, namely in the late Sasanian period.¹⁰ Thus, Zoroastrians could afford to possess slaves during this time. Following the Arab/Muslim conquests of the seventh century ce, the common assumption is that Zoroastrians lost all their worldly privileges, including the right to own slaves. As a result, the occasional references to slaves in later (post-Sasanian) Zoroastrian literature have been viewed as anachronistic, serving rhetorical purposes only. In his commentary on such references in a late Pahlavi text conventionally known as The Pahlavi Rivāyat Accompanying the Dādestān ī Dēnīg (PRDd), Alan Williams writes:

There are several references in PRDd. to matters which would have been anachronistic in the 9^th–10^th centuries A.C. under Muslim rule … It is doubtful also whether in this period Zoroastrians would have been able to own slaves as they had in former times. The mention of slaves in 34c2 has a rhetorical function.¹¹

While mentions of slaves in this text and other similar Pahlavi and Persian works reflect the learned traditions of Zoroastrian priests (see below), it should not be assumed that Zoroastrians in post-Sasanian times were unable to own slaves. It was certainly not the case for late medieval and early modern Parsis, yet the evidence supporting this has not been given serious consideration. Beside the fact that the later period of Zoroastrian history has traditionally been neglected by scholars, we can identify one reason for this situation: the possession of slaves is seen as conflicting with the dominant narrative of oppressed, marginalized, and impoverished medieval Zoroastrians who lived under the rule of foreign (often Muslim) authorities in both Gujarat and Iran.¹² This presumed marginalized and impoverished status begins to dissipate when Parsis started to prosper in the late seventeenth and eighteenth centuries. During this period, as the Parsis’ economic fortunes significantly improved through trade, some of them were able to employ domestic “servants” in their households.¹³

Indeed, the narrative of oppression and hardship faced by medieval and early modern Zoroastrians, especially Iranians, is partly true, and we do find reflections of it in certain passages in the Persian Rivāyats (see Conclusion). However, it is just one aspect of the story and should be complemented by other pieces of evidence that provide a more balanced view.

This paper aims to highlight instances in which Zoroastrians of this period spoke about their own slaves. First, it seeks to demonstrate that the passages in the Persian Rivāyats reflect the problems faced by Zoroastrians, particularly Parsis, with slaves since the late fifteenth century. Second, it examines the diverse reactions of both Iranian and Parsi communities regarding the question of converting slaves to Zoroastrianism. Before delving into an analysis of the relevant passages, it is necessary to provide details of the nature of the Persian Rivāyats. This will be followed by a discussion of the broader context of slave ownership in Gujarat and Iran. Subsequently, the traditions of Zoroastrian priests regarding slaves and slavery will be outlined, drawing on Pahlavi and Persian texts.

Persian Rivāyats

Although they do attest to the existence of slavery and the slave trade in the societies where Zoroastrians lived, contemporary European accounts are not very informative regarding this aspect of Zoroastrian life.¹⁴ Our main source of information regarding slaves during this period is a collection of letters known as the Persian Rivāyats.¹⁵ These letters consist of a series of occasional correspondences in Persian between Iranian Zoroastrians and Parsis from the late fifteenth to the eighteenth century.

In these letters, Parsi communities (or rather Parsi travellers to Iran) ask technical questions about various aspects of the religion from Iranian Zoroastrians, who act as religious authorities in these exchanges.¹⁶ Each individual letter is usually named after the messenger to and from Iran (e.g. Rivāyat of Narīmān Hūshang) or an esteemed Parsi priest or layperson who is addressed in the letter (e.g. Rivāyat of Borzū Kāmdīn). During the seventeenth century, some Parsi priests began to collect these letters from Iran and used many of them as the primary material for the compilations they put together. It is in this compiled form that we have access to most of the earlier letters.¹⁷ The flow of correspondence did not cease after these compilations but continued until 1773 when internal divisions within Parsi communities over the calendar and other religious issues led to the cessation of correspondence with Iranians. Both these later letters and the earlier compilations were copied and transmitted by successive generations of Parsi priests.

The internal structure of the letters can be generally described as follows:¹⁸ they typically begin with invocations addressed to the Parsi addressees, both lay and priestly. This is sometimes followed by a description of the current situation of Iranian Zoroastrians. Subsequently, the letters usually present the Parsi questions alongside the corresponding Iranian answers, referred to as “porsesh va pāsokh-e dini” or “porsesh-e dini” (religious question and answer, or religious question). These inquiries primarily revolve around various religious topics such as marriage, death ceremonies, purity rituals, prayers, doctrinal matters, and more, in a loosely organized manner. References to domestic slaves occasionally appear in this central part of the letters. Finally, they are often concluded by providing a list of the names of Iranian priests and laypeople who contributed to the answers.

The core of these letters, consisting of questions and answers, resembles the older genre of question and answer in Pahlavi literature. In this genre, a series of socio-religious questions was posed to a religious authority.¹⁹ Some questions appear to be hypothetical/traditional and may not necessarily reflect a concrete problem within the Zoroastrian community or the individuals who ask them. The priests typically respond to the questions based on what is commonly referred to as dēn (tradition, religion) or avesta (sacred texts). In the case of the Persian Rivāyats, the religious authorities were Iranian priests. Their answers sometimes contain earlier materials found in priestly traditions, particularly in subjects related to purity laws. As a result, both in terms of style and content, the Iranian responsa (and even some Parsi questions) demonstrate clear connections to earlier Pahlavi and Persian texts. Recognizing this connection, however, carries the potential danger of relegating the Persian Rivāyats to the status of ancillary texts primarily useful for understanding earlier traditions, which may lead to overlooking the wealth of information they provide about lived Zoroastrianism.²⁰ A considerable number of questions in these texts reflect the challenges and concerns faced by the Parsis in their own time. Consequently, the answers provided by the Iranian priests often incorporate new elements arising from the novelty of the circumstances. This is because the solutions could not be derived solely from earlier instructions, prompting the priests to draw inferences from the tradition.²¹ In summary, the questions and answers found in the Persian Rivāyats should be approached primarily as valuable sources shedding light on the contemporary history of both the Parsi and Iranian communities, rather than solely as a means of understanding earlier traditions.

Historical and socio-cultural context of slavery in Gujarat and Iran

Gujarat, during the period under consideration, was initially ruled by the local Muslim sultanate of Muzaffarids. Subsequently, in the sixteenth century, parts of their territory came under the control of the Portuguese and parts under the rule of the mighty Mughals. This balance between the Mughals and Portuguese was interrupted during the seventeenth and eighteenth centuries by the arrival of other European powers, notably the Dutch and English East India companies. Additionally, the Hindu Marathas began to rise, further altering the dynamics, ultimately leading to the decline of Mughal influence. It is widely believed that prior to the seventeenth century, Parsis predominantly lived as agriculturalists, leading secluded lives in Gujarati villages.²² It follows that although Parsis enjoyed a great deal of tolerance in India, they were not initially wealthy. It was only with the arrival of Europeans (more specifically British) that they embarked on their trading endeavours and subsequently achieved prosperity.²³ However, there is evidence suggesting that even before the arrival of Europeans in Gujarat, some Parsis were already engaged in trade.²⁴ For instance, a Parsi merchant named Čāhil Sāngan, who resided in Cambay in the fourteenth century, was rich enough to invite an Iranian priest to Gujarat and provide him with financial support for the copying of several Avestan and Pahlavi manuscripts.²⁵ Moreover, many of the Parsi messengers mentioned in the Persian Rivāyats seem to have been merchants. For example, a messenger named Esfandīyār Sohrāb, who travelled to Iran around 1520, likely had economic interests and connections in the trading hubs of the Persian Gulf.²⁶ In addition to merchants, the ranks of wealthy Parsis also included large landowners, exemplified by figures like Chāngā Āsā, a distinguished lay community leader in Navsari during the fifteenth century.²⁷

On the other hand, during the fifteenth century, Iran and Central Asia were divided among the descendants of Timur, Turkman dynasties, and local chieftains. In the early years of the sixteenth century, Shah Ismail (r. 1501–24), the son of a Sufi leader, successfully centralized Iran both politically and religiously. He established the Safavid dynasty, which remained in power until 1722. The eighteenth century is marked by political instability in Iran, particularly in regions with a dense Zoroastrian population like Kerman and Yazd. Periods of stability were rare and intermittent, mainly under the rule of Nader Shah Afshar (r. 1736–47) and Karim Khan Zand (r. 1753–79).

The portrayal of Iranian Zoroastrians of this period in modern literature depicts them as even poorer than their Parsi counterparts – a marginalized and suppressed community. They lived as petty farmers in cramped, fortress-like houses, toiling on the fields of remote villages in Yazd and Kerman.²⁸ Although it cannot be denied that many Iranian Zoroastrians were poor farmers, scattered references in both the Persian Rivāyats and European accounts suggest the existence of merchants and wealthy individuals among Yazdis and Kermanis. A letter dated to 1511 was carried to India by three Yazdi Zoroastrians who were travelling there for commercial purposes (az jahat-e tejārat).²⁹ Jean-Baptiste Tavernier, a French merchant and traveller of the seventeenth century, speaks about the involvement of Kermani Zoroastrians in the lucrative trade of Kerman wool.³⁰ Additionally, a short versified text in Persian called Qesse-ye Kāvūs va Afsād, likely composed in the sixteenth century, narrates the story and challenges faced by two Zoroastrian merchants from Yazd as they embark on a commercial journey to India.³¹

Gujarat, with its thriving ports, was one of the most important nodes in an extensive maritime trade network in the Indian Ocean. This network encompassed the western and eastern coasts of India, the Persian Gulf, eastern Africa, the Red Sea, and Southeast Asia.³² One of the commodities exchanged between these trading nodes was slaves with diverse origins. It is well-known that a significant number of these slaves were imported to Gujarat from eastern Africa, particularly from Ethiopia, giving rise to the prominent African slave population known as Sidis or Habashis. These Sidis, often serving in the armies of Indian states, eventually gained considerable political and military influence in late medieval and early modern Gujarat, as well as in other regions of India.³³

However, it is crucial to note that not all slaves in Gujarat were imported from Africa, and their roles extended beyond military service. Many Indians themselves could be enslaved or find themselves in situations similar to slavery due to factors such as famine, extreme poverty, debt, or belonging to certain lower castes.³⁴ European accounts, which grew in number from the seventeenth century onwards, paid close attention to these lower-caste and outcaste groups in Gujarat and the services they provided to the higher castes. In these accounts, the outcaste Indians in Surat were collectively referred to as halālkhors (lit. “those who eat halāl food”). They were assigned menial and polluting tasks, such as sweeping houses, cleaning streets, handling waste and dung, washing dead bodies, and carrying them to appropriate locations for Muslims, Parsis, and upper-caste Hindus. As a result, the latter groups considered halālkhors to be extremely impure and avoided any form of contact with them.³⁵

The existence of various forms of slavery in late medieval and early modern Iran is well-documented through official manuals, documentary sources, European accounts (particularly from the Safavid period onwards), and theological works pertaining to slaves. The ports of the Persian Gulf, as well as the northern and western borders, played a significant role in the importation of slaves into Iranian territories. As enslaving fellow Muslims was theoretically illegal, many male and female slaves (gholāmān and kanizakān) were obtained either from the Christian population in the Caucasus or through the Persian Gulf trade and overland Hajj route from among East Africans and Indians.³⁶ During this period, the Shia clergy continued the tradition of composing treatises concerning slaves, especially in a genre dedicated to their manumission known as ʿetq. One notable example is a treatise called ādāb-e saniyeh, authored by a certain Mohammad Saleh Khatunabadi in 1708. This work, patronized by Shah Sultan Hossein, the last Safavid king (r. 1694–1722), delves into various aspects of slavery, including the rules for manumission, the sources of slavery, and the relationship between slaves and their masters.³⁷ European accounts of the time not only documented the possession of slaves by the Muslim majority in Iran but also recorded the case of wealthy Armenians who owned slaves.³⁸ Tavernier provides an interesting account of the life of an African slave who belonged to an Armenian merchant from New Julfa. This merchant had close ties to Shah Safi (r. 1629–42). His agents had purchased the slave from the port of Malindi, located on the Swahili coast. Despite being initially converted to his master's religion, Christianity, the slave later embraced Islam after his master's tragic suicide. He spent almost two decades as a wandering dervish, travelling across various regions of the Middle East. However, it seems that Islam failed to fulfil his spiritual aspirations. Eventually, he returned to New Julfa, burdened with shame for his apostasy. To repent, he adopted an extremely austere lifestyle, earning respect within the Armenian community.³⁹

Slaves and slavery in Zoroastrian priestly traditions

This section briefly reviews the priestly traditions on slaves. These traditions are manifested in texts composed and transmitted in Avestan, Pahlavi, and New Persian. As is valid for many other subjects, the early texts in Avestan provide limited information about slavery. This scarcity has led some scholars to propose that the institution was not inherently Iranian/Zoroastrian, but rather an alien institution that later Zoroastrians adopted through their interactions with neighbouring cultures.⁴⁰ Our knowledge of the social structure of the Avestan people is imperfect. However, based on terms like vaēsa- (meaning “servant”), we can assume the existence of a certain form of strong asymmetrical dependency among the Avestan people.⁴¹ Even if we accept that slavery did not exist among these people, this absence does not make the institution any less Zoroastrian/Iranian.

Pahlavi texts, composed and redacted during the late Sasanian and early Islamic period, provide abundant information on slaves, in contrast to the limited references found in Avestan texts.⁴² These works clearly demonstrate that there was no moral prohibition against owning slaves. A Pahlavi exegesis (zand) of a section of Avestan Hērbadestān incorporates discussions about slaves within the context of the sinful act of returning the children of non-Zoroastrians/sinful Zoroastrians to their families. According to this text, a non-Zoroastrian slave who converts to Zoroastrianism is freed from the ownership of their non-Zoroastrian master. It is also prohibited to sell a Zoroastrian slave to non-believers.⁴³

The most comprehensive Pahlavi text concerning slaves is Mādayān ī Hazār Dādestān, a compilation of legal cases compiled by a figure named Farroxmard ī Wahrāmān. This work comprises actual and hypothetical legal scenarios covering various subjects from diverse sources.⁴⁴ It provides valuable insights into the legal and social status of slaves in late Sasanian society. A whole chapter addresses the legal regulations surrounding slavery, while references to the institution are dispersed throughout the work. According to Maria Macuch, the primary sources of slavery during the late Sasanian era were war captivity, being born into slavery, debt bondage, and extreme poverty leading to the sale of one's freeborn children or wife. These enslaved individuals were primarily employed in three sectors. A significant number served as domestic slaves within their masters’ households, responsible for all menial tasks. Others were engaged in labour within large or small agricultural units. They legally belonged to the land and would be transferred to a new owner if the land was sold. Another category of slaves also existed, namely those who worked on lands owned by fire temples. Known as anšahrīg ī ātaxš (“slave of the fire”), these slaves could be non-Zoroastrians.⁴⁵

It must be noted that slaves could come from various backgrounds, and were not always of foreign origin. They could even be Zoroastrian. In line with the Pahlavi exegesis in Hērbadestān, MHD emphasizes that the Zoroastrian slaves should only be owned by Zoroastrian masters and must not be sold to non-believers. Violating this rule would render both the seller and purchaser guilty and subject to legal punishment.⁴⁶ This rule, with some variations, is reiterated in later texts as well. In its post-Sasanian version, as attested in PRDd, apart from facing worldly punishment exerted by Zoroastrian authorities, both the seller and purchaser would bear responsibility for any sin committed by the slave after his conversion away from Zoroastrianism.⁴⁷ Chapter 30 of Saddar-e bondahesh, a Persian text from the thirteenth or fourteenth century ce, does not mention any worldly punishment for wrongdoers. Instead, only the Zoroastrian seller would be considered morally culpable, and the subsequent sins committed by the slave would be attributed to his account.⁴⁸ Another statement in MHD that favoured Zoroastrianism stated that a slave belonging to non-Zoroastrians, upon converting to Zoroastrianism, had the right to buy his freedom and instead become a subject of the king.⁴⁹

Like domestic animals and immovable objects, slaves were considered as property (xwāstag), according to MHD. Consequently, they could be used in various transactions. In other words, they could be given as gifts, shared by two free individuals, leased, and so on. Despite their status as property, slaves had certain limited forms of subjectivity, either granted by their owner(s) or by law. Owners could grant them the right to acquire property or allow them to receive gifts from third parties. Additionally, despite some reservations, slaves could participate in legal proceedings as either one of the parties or as witnesses. They also had the right to take their masters to court if they were mistreated. Manumission was possible through the will of the owner, resulting in full or partial freedom. Partial manumission would have occurred when the slave was jointly owned by two or more masters. In such cases, one of the owners would release the slave based on their proportionate share of ownership.⁵⁰

Apart from MHD, there are occasional mentions of (domestic) slaves in other priestly works written in Pahlavi, Pāzand (Pahlavi transcribed in Avestan script), and New Persian. In these texts, slaves are depicted as integral members of an ideal Zoroastrian household, along with the head of the family, wife, children, domestic animals, and the hearth fire. On the one hand, the wisdom literature advised the master of the house and the children to treat slaves with kindness and avoid mistreating them.⁵¹ It was also the responsibility of the master to perform or arrange necessary religious ceremonies for the soul of a deceased slave.⁵² On the other hand, the same traditions emphasized the importance of slaves’ obedience and diligent service to their masters. If a slave failed to obey, the master was granted the authority to use physical force to restore obedience.⁵³

The literature reviewed here illustrates that Zoroastrian priests held traditional discourses concerning slaves. These traditional discourses and their underlying logics likely exerted some influence on the content of the passages in the Persian Rivayats concerning slaves, especially in the responses provided by the Iranian priests. To find out more, we should now examine these passages.

Domestic slaves in the Persian Rivāyats

It is only in four out of more than 25 letters that domestic slaves are explicitly mentioned in the Persian Rivāyats. These letters are the second letter of Narīmān Hūshang, Kāmā Bohrā, Kamdin Shapur, and Ithoter Rivāyats. The first three Rivāyats date from the fifteenth and sixteenth centuries, while the latter dates from the eighteenth century. In the following analysis, we will examine these passages in chronological order.

The second letter attributed to Narīmān Hūshang is dated to 1480/86.⁵⁴ It contains Iranian responses to a set of questions posed either by Parsis in Navsari or by a Parsi messenger. The letter is signed by a group of priests from Sharifabad and Torkabad, two villages located on the northwestern edge of the Yazdi plain and known for their significant number of Zoroastrian priests.⁵⁵ The addressees of the letter were exclusively the prominent Parsis in Navsari and did not include individuals from other Parsi settlements mentioned in Narīmān's first letter (dated 1478). One of the addressees was a distinguished lay Parsi named Chāngā Āsā, who is described as the leader of the town of Navsari (sālār-e shahr-e nowsāri). It appears that he played a prominent role in the internal affairs of the Parsi community in Navsari and its neighbouring areas. He served as the representative of the community in its interactions with Muslim authorities. In this context, Iranian priests behind the first letter of Narīmān Hūshang praised his ability to exempt the Parsis of Navsari from the poll-tax (jeziya) imposed on non-Muslims by the sultans of Gujarat.⁵⁶ We gain further information about him and his other pious activities from a late sixteenth-century poem called Qesse-ye Sanjān, which narrates the story of the migration of the Parsis from Iran to the shores of India. Among Chāngā Āsā's notable achievements was the initiative to transfer the Iranshah fire, the only Bahram fire of the Parsis at that time, to Navsari.⁵⁷ In addition to him, the letter was also addressed to his son Bahram, Hērbad Rostam, Hērbad Hūshang, and Hērbad Khorshīd, the head of the Sanjana lineage of priests in Navsari. Through this correspondence, Iranians were informed that the Parsis did not observe certain rituals and practices in the same manner as they did.⁵⁸

The subject of interaction with non-Zoroastrians takes centre-stage in this letter.⁵⁹ In this context, the question of whether domestic slaves could be converted to Zoroastrianism was asked:

(Question: regarding male and female slaves who have faith in the Good Religion, is it obligatory to tie the sacred girdle around their waist?)

Answer: if Zoroastrians trust them, it is obligatory to tie the sacred girdle around their waist. And when they become wise, knowledgeable in religion, and steadfast, Zoroastrians should perform barashnūm for them. It is also obligatory and permissible (for the laity) to eat something from their hand. (However, the case is different for priests. It is not allowed until three generations (korsī)⁶⁰ have passed.)⁶¹

The terms used here to refer to domestic slaves were gholāmān and kanizakān, which were the conventional words for male and female slaves in the Persian-speaking world at that time. It is noteworthy that archaic terms such as anšahrig, bandag, or paristār were not used in this passage. This question-and-answer reveals that as early as the late fifteenth century, affluent Parsi families had already employed slaves to serve in their homes. This challenges the widely held assumption among scholars that Parsis only started employing domestic slaves from the seventeenth century onwards, as their economic status improved.⁶²

The reason(s) behind the Parsis’ inquiry regarding the conversion of their domestic slaves remains unclear. From the formulation of the Parsi question, it appears that the slaves themselves might have expressed a desire to embrace the Zoroastrian religion.⁶³ The question, based on the latter part of the Iranian answer, could also indicate Parsi concerns about purity. It is likely that certain Parsis, especially those behind the question, felt uncomfortable with the idea of consuming food prepared by these non-Zoroastrians, due to their perceived ritual impurity. According to Zoroastrian prescriptions, it was forbidden to consume food prepared by non-Zoroastrians. In my view, the Iranian response appears to be driven more by the priestly concerns of its authors than anything else. Another plausible reason behind posing the question is that certain Parsis might have had a sincere interest in the conversion of their slaves, stemming from the genuine concern for their salvation. In addition to the question about converting slaves, the Persian Rivāyats corpus also includes other Parsi inquiries related to conversion. These questions pertain to the reconversion of apostates and the general possibility of converting non-Zoroastrians.⁶⁴

The response given by Iranian priests was positive in nature. If Zoroastrian masters trusted their slaves (be behdīn eʿtemād bāshand), the latter could be converted. The process of conversion consisted of the following steps: after receiving religious instruction and becoming proficient in essential prayers, they could proceed with the elaborate purification ritual known as barashnūm to be purified from their inherent pollution. Only after this ritual, compulsory for anyone converting to Zoroastrianism, were they allowed to wear the sacred garments.⁶⁵ After these steps, they were considered Zoroastrian and the laity were allowed to partake of the food prepared by these slaves. Priests were bound by more stringent regulations regarding purity and pollution. They were strictly prohibited from consuming such food. Only members of priestly families who had not undergone priestly initiation ceremonies within the past three generations were permitted to partake in this type of food. The situation was more lenient for the latter, as male descendants of priests lost their eligibility to pursue the office of priesthood after three generations without undergoing a new priestly initiation ceremony.⁶⁶

The next mention of slaves can be found in the Rivāyat of Kāmā (Asā) Bohrā, dated to 1527.⁶⁷ Although named as such in the classified Rivāyat of Dārāb Hormazyār, it is more likely that the messenger of the letter to India was Shāpūr Āsā, probably the brother of Kāmā Bohrā. Unfortunately, little is known about these two people, except that they were merchants (bohras, a Gujarati term denoting merchants) from Cambay.⁶⁸ In the fifteenth and sixteenth centuries, Cambay was the most significant entrepôt in Gujarat, boasting extensive trade networks that reached across both west and east Asia, and had strong connections with the Portuguese. Kama Bohra's Rivāyat is one of the longest letters, dealing with a diverse range of subjects that were asked by the Parsis in Navsari and Cambay. Unlike earlier letters, we have access to the original Iranian letter written by Dastur Shahrīyār Ardeshīr Iraj Rostam, a priest-scribe with Khorasani roots, who resided in Yazd at the time of composing the responses. It should be noted that the content of this letter is sometimes similar (if not identical) to another Rivāyat attributed to Kāvūs Kāmān. This has led to confusion between the two in the Dārāb Hormazyār Rivāyat.

Among the various subjects discussed in this letter, one question-and-answer addressed the question of a slave participating with his master in the recitation of prayers during a religious ceremony. This ceremony should have taken place in a domestic setting:

(Question:) that there is a man who takes a bāj (bājgīr), and he has a slave. Is it proper when the master, who is taking the bāj, takes the bāj (Avestan formula) from his slave?

Answer: if he has manumitted the slave and the latter has the appropriate sacred girdle and shirt and recites the bāj correctly, it is proper. However, if everything is in order but the master has not manumitted the slave, it is not proper for the master to take the bāj from the slave.⁶⁹

This passage requires some background information. The term bāj (also spelled bāzh, pāzh) refers to a series of prayers and formulas in Avestan, the sacred language of Zoroastrians. These prayers are recited before, during, and after specific actions, be they mundane or religious. They serve as a spiritual boundary, containing impure activities and upholding the purity of meritorious acts. Thus they framed a wide range of activities, from performing the sacred ritual of yasna to consuming a meal. During the action itself, strict silence must be maintained, and only after completing the final bāj one can resume normal speech. Originally, in ancient and medieval times, the recitation of bāj for various purposes was performed by all adult Zoroastrians, both men and women. However, in modern times, their use has primarily been limited to priests, both in India and Iran.⁷⁰ The term bājgir used in our passage can be interpreted in two ways: it might designate a practising priest who regularly recites the bāj,⁷¹ or it could refer to any Zoroastrian man reciting such a formula.

One of the most common occasions for reciting bāj is before and after daily meals, a practice commonly known as bāj-e nān khordan (the bāj of eating a meal). The prayers recited during this bāj can vary in length, reflecting the knowledge of the reciter or the significance of the occasion. The most elaborate version of bāj-e nān is called drōn yašt, which includes reciting a liturgy (mainly yasna 3–8) and performing a series of ritual actions. Drōn yašt could be performed as either an independent ceremony or be incorporated to other, more elaborate rituals. It could be administered either collectively or individually. In its collective version, the recitation of prayers is divided among two or more participants, sharing the responsibility of the recitation. In pre-Islamic times, there is clear evidence that both the priesthood and the laity used to participate in the performance of drōn yašt.⁷² The ceremony implied in our passage probably refers to a joint drōn yašt ceremony involving both a slave and his master.

It seems that both the Parsi question and Iranian answer belong more to the learned tradition than addressing an actual problem of the time. In medieval and early modern times, the celebration of drōn yašt was primarily associated with priestly functions. Nevertheless, this passage indicates an awareness of the possibility of laymen engaging in such rituals. The Parsi question is whether the slave could participate in the shared ceremony of drōn yašt. According to Iranians, for the participation in any ritual, meeting the general criteria, namely possessing knowledge of the proper prayers and wearing religious garments properly, was essential.⁷³ But the Iranian priests maintained that slaves were not permitted to have any role in such rites with free men, even if they met other criteria. The necessary condition for their participation was their freedom.

The next letter mentioning slaves was brought to India by Kāmdīn Shāpūr in 1558. He came from a priestly family in Broach. In addition to the letter, he also brought some manuscripts and ritual objects to India. The letter was primarily addressed to the Zoroastrians of Broach, especially to Dastur Padam Rāmyār, the head of the priests there. Thus it is reasonable to assume that this letter was a response to questions posed by the Zoroastrian authorities in this port.⁷⁴ It contains the names of priests and laity from Torkabad, Sharifabad, Khorasan, Sistan, and Kerman.

In this Rivāyat, there is a passage that may refer to the conversion of female slaves. Regrettably, the Parsi question has not been transmitted to us. Nevertheless, we do possess the Iranian response, which elaborates on the necessary procedures that must be followed for converting them:

A young female slave (has been) purchased (bardeh-ye kharīd-e kūchak?).⁷⁵ If she has not yet menstruated, she should be fed for three nights in lard. Then her head should be washed with barashnūm. If she has already experienced menstruation, she should be fed for 41 nights in lard. Then her head should be washed with barashnūm. Thereafter she should be given in marriage. ⁷⁶

The conversion process described here corresponds with parts of Anquetil-Duperron's description of conversion to Zoroastrianism quoted above. Providing food to the candidate for a certain number of days is also attested by him, as he mentions that for three days the candidate is given pure food prepared by Zoroastrians. Zoroastrians assumed that the food consumed by the candidate up to that moment was impure. While Anquetil's focus was on the general description of the conversion rituals, our passage focuses on the conversion of women. As with any other subject concerning women, menstruation plays a key role in the process. If the acquired female slave is pre-pubescent, she is given pure food for three nights in a space called lard. Following this period, she would undergo an initiatory barashnūm. If the female slave was already an adult (and hence had already menstruated), she was considered more seriously polluted, and the duration of giving her pure food was longer. Only after 41 days could she undergo a barashnūm. Moreover, as almost all adult Zoroastrian women were married, this newly converted woman should also be given in marriage to a Zoroastrian man.⁷⁷ Therefore, Iranian priests were open to the idea of converting and marrying these women. Nevertheless, the passage lacks detail concerning the legal consequences of such unions for the slave woman and the children born to her. It is very likely that after her conversion her legal status would have changed for Iranian priests. She would have been manumitted and integrated, along with her future children, into the Zoroastrian family and community.

A few words must be said regarding the ambiguous lard. In her fascinating account of the religious life of the Zoroastrian village of Sharifabad during the 1960s, Mary Boyce described a structure known as lard in the following manner: “a small square mud-brick building, on the outskirt of the village [Sharifabad], with a chimney-like hole in the flat roof, and steps leading up to this, but no door”. This building was used to dispose of impure items, such as trimmed hair, nails, and the corpses of prematurely stillborn children. On occasion, acid was poured into the lard through the hole to dissolve the waste inside.⁷⁸ The relationship between this building and the lard mentioned in our passage is unclear, however. Drawing on Persian lexicons, Dhabhar translated the word as “public place”.⁷⁹ In the description of Yazd's monuments and urban spaces by Iraj Afshar, we encounter a similar meaning. According to Afshar, lard denotes a square-like space in the old Persian towns where goods intended for use by the townspeople, and brought from neighbouring regions, were deposited.⁸⁰

After Kāmdīn Shāpūr's Rivāyat, there is a noticeable absence of slaves in the subsequent correspondence between the Parsis and Iranians until they reappear in the second half of the eighteenth century in the Ithoter Rivāyat. The Ithoter Rivāyat (Guj. “seventy-eighth Rivāyat”) marks the final exchange of letters between the two communities, and dates to 1773. A unique Gujarati document from the eighteenth century sheds light on the subject of slaves. This document, a deed of partition, attests to the presence of both female and male slaves in the household of Nowrozji Kersaspji Homji Unvala, a wealthy Parsi from Surat. The document, beside the distribution of other properties, assigns the slaves among Nowrozji's heirs, with some inheritors sharing joint ownership of certain slaves. Interestingly, the origin of these slaves is traced back to a low-caste Indian group known as Kolis. In his insightful analysis of this valuable document, Jivanji J. Modi, a renowned Parsi scholar, referred to the practice of employing Kolis by large landowners (zamīndārs) in early modern Gujarat. He further elaborated on the duties and rights of these Kolis in their relationship with their masters. The masters were responsible for providing food and care. In return, Kolis had to serve in their fields and homes. Their servile status was hereditary, passing on to their children, who also had obligations towards their parents’ masters.⁸¹

In the eighteenth century, the Parsi community was embroiled in several disputes, including a conflict between the Qadīmī and Shāhānshāhi factions. The division of the Parsis into Qadīmīs and Shāhānshāhis was the result of differing responses to the crystallization of a one-month difference between the religious calendars of Iranian and Indian Zoroastrians during the first half of the eighteenth century. The Qadīmīs adhered to the religious authority of Iranians and decided to follow their calendar, while the majority, Shāhānshāhis chose to remain loyal to the traditional Parsi calendar.⁸² Apart from the subject of calendars, this period also witnessed disagreements over other religious matters, including those related to domestic slaves. The questions asked in the Ithoter Rivāyat reflect these internal conflicts from the perspective of the Qadīmīs. The port city of Surat was one of their primary strongholds, where the British East India Company (EIC) had substantial influence since 1759.⁸³

The messenger chosen for delivering the 78 questions to Iranians was Mollā Kāvūs, son of Rostam, coming from a priestly family in Broach. His voyage to Iran was sponsored by Dhanjishāh Manjishāh, a lay merchant and fervent supporter of the Qadīmīs, who also acted as the agent of the EIC in Surat. Dhanjishah found himself entangled in a rivalry with another prominent lay merchant, Mancherji Khurshedji Seth, who had affiliations with the Dutch East India Company and backed the Shāhānshāhis. The two merchants engaged in a kind of power struggle, each seeking to exert influence over the internal affairs of the Parsis in Surat.⁸⁴ Mollā Kāvūs managed successfully to reach Yazd, where he presented the questions to an assembly of Zoroastrians there. It is commonly assumed that Iranian Zoroastrians were enduring challenging circumstances during the turbulent eighteenth century in Iran.⁸⁵

Out of the 78 questions, two specifically addressed the subject of domestic slaves. In broad terms, these two questions relate to the permissibility of exposing the deceased bodies of slaves in Zoroastrian towers of silence (dakhmehs). This issue foreshadows the better-documented disputes among modern Parsis, wherein the bodies of sinful or polluted Parsis were occasionally denied placement in dakhmehs. Instead, their bodies were laid in a separate enclosed structure known as a chotra, serving as an alternative to the towers of silence.⁸⁶

The thirteenth chapter of Ithoter Rivāyat provides the most comprehensive account of slaves in the collection of the Persian Rivāyats. It offers insights into the complex treatment of slaves by the Parsi community. The Qadīmīs framed their question in a way that highlighted the paradoxical behaviour of the majority of Parsis, presumably the Shāhānshāhis. On the one hand, the latter Parsis show a keen interest in educating Hindu children they purchase,⁸⁷ teaching them Avestan prayers, and providing them with sacred garments. They readily consume the food prepared by these slaves in their everyday lives, and during religious ceremonies and festivals priests consecrate the food. However, a contrasting attitude emerges when these slaves pass away. The same Parsis refuse to expose their corpses in the dakhmehs, considering them the offspring of “infidels”. They argue that the bones of these slaves should not be placed next to those of other Zoroastrians. Therefore, seeking clarity on this subject, the Qadīmīs approached Iranians to inquire about the permissibility of exposing their corpses in the dakhmehs:

Question: in this quarter (India), Lay Zoroastrians of India purchase many boys and girls of Hindus as male and female slaves (be gholāmī va kanīzī) and maintain them in their domestic work and service. They (Parsis) teach them Avesta and have them wear the sacred girdle and shirt in accordance with Zoroastrian rituals. They (Parsis) also arrange and consecrate the drōn-e gāhānbār and other things which they (slaves) have prepared. Water and food are also taken from them by both priests and laity of India. Yet when these slaves die, those priests and laity refuse to allow their corpses to be placed in the dakhmeh, claiming that they are the children of evil ones (drovands) and that it is not proper to put the bones of Zoroastrians with theirs in one place. Thus, while these individuals were alive, they made use of them for all religious preparations, and after their death they (the Parsis) do not allow them to be laid in dakhmeh. The question is, therefore, whether it is proper or not to lay the corpses of these slaves in the dakhmeh. Let them (Iranians) write to clarify this matter.

The answer, given by a consensus of Yazdi priests, is affirmative. They first emphasize that these purchases should not jeopardize the religion, property, or lives of Zoroastrians. This precaution not only highlights the potential dangers of buying slaves in Muslim Iranian society but also underscores concerns about the quality (both physical and mental) of these slaves. If no harm is caused, the act of buying and converting them is considered commendable. Consequently, their bodies must be placed in the dakhmehs as well. Iranians strongly criticized the ambivalent treatment of Parsis, cautioning that they would be deemed deserving of death (margarzān):

Answer: concerning the purchase of non-Zoroastrian (joddīns) boys and girls, the priests and laity must first show care for their own religion, rite, soul, and property so as not to face losses. When that is considered, buying non-Zoroastrian children, teaching Avesta to them, and converting them to the good religion of Mazda worshippers earns one a great merit.⁸⁸ What is instead highly condemnable and non-conforming with the opinion of the members of the good religion is the fact that the priests and laity in India should eat food prepared by those boys while they live, and then once they die and stand to face God's mercy, they should make such base comments about their poor bodies, arguing inappropriately that they are children born from non-Zoroastrians. And that their corpses should not be united with those of Zoroastrians in the same dakhmeh. It is not right! Such unjust arguments do not benefit from the religion of Zarathustra and the righteous path. Whoever behaves in this manner and does not allow their bodies to be laid in the dakhmeh is, according to the religion, worthy of death and accountable before Mehr and Sorūsh.⁸⁹ Indeed, it is necessary for both priests and laity to show greater respect for these children and to allow the bodies of these deceased ones to be placed in the dakhmeh according to the rules of good religion. This will bring joy to Ohrmazd and Amshāspands.

Furthermore, Mollā Kāvūs informed Iranians that in various Parsi settlements, both the priests and laity have passed a regulation (band va bast nemūdan) to prevent the conversion of these children. Interestingly, in the nineteenth century, the Parsi Gujarati derivative bandobast was frequently employed in connection with the resolutions of the Bombay Punchayet.⁹⁰ This band va bast nemūdan may refer to an otherwise unknown resolution regarding slaves that was adopted by the Bombay Parsi Punchayet or a similar communal institution during the eighteenth century:

Secondly, we have heard from the words of the descendant of priests, Dastūr Kāvūs, the worthy successor of the deceased Dastūr Rostam, that many priests (dastūrs and mōbads) and laity throughout the country stand in the way. And they are an impediment and have issued resolutions not to teach those boys Avesta and not to convert them to the good religion of Mazda worshippers. This is unreasonable and alien to the tradition. May the beloved ones prosper! In the second chapter of Juddīvdād (Vīdēvdād), the creator of the righteous material world has ordered the honorable Zarathustra Esfantamān, may his soul be blessed, to guide all men to the religion of goodness, to the main path, to edify his joy, his glory, and his honour. Secondly, during the time of Hōshīdar Māh, Hōshīdar Bāmī and Sōshāns, all non-Zoroastrians will be converted to the good religion. It follows that according to the religion of goodness, it is appropriate and necessary to convert these boys, as it is a very great merit and a virtuous act. Therefore, those who hinder this can be considered as followers of non-Zoroastrian religion. And they are not even aware of the origin and resurrection. They proceed along the path of deviation and vanity. According to the religion, they cannot be called Zoroastrians, for if they were, they would have contributed to the growth of the good religion.⁹¹

To justify the conversion of these slaves, Iranians put forward two arguments. First and foremost, they cited the second chapter of Vīdēvdād, an Avestan text focusing on purity laws and the eradication of evil forces, as a scriptural authority. According to their interpretation of this chapter, Ohrmazd had instructed Zoroaster to convert all of humanity to Zoroastrianism.⁹² Their second argument revolved around future events. Iranians reminded the Parsis about the three future saviours, who were believed to be the sons of Zoroaster, and their ultimate mission to convert all non-Zoroastrians to Zoroastrianism. They concluded that those who opposed the conversion of these slaves were unworthy of being called true Zoroastrians, as a genuine follower of the faith actively propagates it. In this way, Zoroastrianism was presented as a universal religion.

In this important question and answer, three distinct approaches to the subject of converting Hindu slaves can be identified. The first approach, supported by Iranians and presumably the Qadīmīs, advocates for accepting these slaves as full-fledged Zoroastrians with all the rights and privileges that would come with conversion. As a result, they ought to be exposed in the same dakhmehs used by the rest of the Zoroastrian community.

The second approach stands in stark contrast to the first. It is represented here by certain Parsis, whose viewpoint was conveyed to Iranians by Mollā Kāvūs. These Parsis preferred to exclude these slaves completely from the fold of Zoroastrianism, both during their lifetimes and after their deaths. According to them, the non-Zoroastrian origin of these slaves (or rather their low social status) rendered them ineligible for acceptance as Zoroastrians.

If we are willing to acknowledge that there is some truth in the ambivalent behaviour of certain Parsis represented in the Qadīmīs’ question, we can also identify a third approach. This attitude takes a middle ground between complete inclusion and exclusion. These Parsis did convert their Hindu slaves during the slaves’ lifetime. However, this conversion was not aimed at sincerely persuading them to adopt Zoroastrian beliefs, nor was it intended to promote the religion among humanity. Instead, it was primarily driven by ritualistic concerns regarding the impurities that their presence could have caused in Parsi households.

During their lives, these converted slaves were likely treated as lower in status or, at the very least, regarded as different from those who had inherited the religion from their parents. As a result, they would have experienced relative marginalization in Parsi communities. This marginalization is underscored in our passage by their being denied exposure in dakhmehs after their death.

In the same Rivāyat, another question was asked concerning the issue of converting Hindu boys with brands on their bodies and whether, after their death, their corpses could be placed in the towers of silence:

Question: they purchase Hindu boys (pesarān-e hendūvān) as slaves (dar gholāmī kharīd mī-namāyand), and these boys have brands (dāgh) on their bodies due to some harm done to them (be sabab-e āzāri).⁹³ Is it permissible to convert them to Zoroastrianism and have them wear sacred shirt and girdle? Furthermore, when they die, can they be placed in the tower of silence?

Answer: in the case of children of non-believers, when they have brands on their bodies, they can neither be bought nor converted. If someone buys branded children and converts them to the religion, it is considered a grave sin. ⁹⁴

The exact nature of this brand on boys remains unspecified. It is also uncertain whether it was connected to their enslavement or predated it. If these boys were indeed passive victims of such branding during enslavement, as implied by the phrase “be sabab-e āzāri”, it is plausible to assume that the marks were used for their identification or as a form of punishment. This practice was prevalent in both the pre-modern Middle East and South Asia with regard to slaves, often employing a hot iron for branding.⁹⁵ The mention of boys, while excluding girls, raises the suspicion that this brand might be related to a procedure performed on the male reproductive organ. In this connection, one may consider the possibility of castration or circumcision

In stark contrast to their previous pro-conversion answers, Iranian priests firmly rejected the validity of such transactions and conversions. They even went so far as to state that if Parsis were to purchase and convert these children, it would be considered a grave sin. Their negative judgement seems to be rooted in the belief that any form of branding or deformation would render the body extremely impure.⁹⁶ Based on this principle, Zoroastrian bodies must remain untainted and intact. The second part of the Qadīmīs’ question remained unanswered, but from the prohibition on buying and converting these boys, it follows that they should not be placed in dakhmehs either.

Conclusion

Between the years 1630 and 1632, Gujarat and its neighbouring regions faced the harrowing impact of a devastating famine that led to the loss of countless lives. Death was not the only price paid by the Parsis.⁹⁷ In a slightly later letter, dated 1635, a Kermani priest conveyed distressing news to his fellow Parsis in Gujarat regarding the enslavement of a young Parsi man named Shāpūr. Coming from the famine-stricken Gujarat, Shāpūr found himself enslaved to a non-Zoroastrian man in Fars province.⁹⁸ The Kermani priest relates the courageous journey of a Parsi man named Pashūtan to Kerman in a bid to rescue Shāpūr. Pashūtan sought the assistance of Kermani Zoroastrians, and together they embarked on the pious mission to free him from slavery. Following negotiations with the owner and payment of the required ransom, they successfully freed Shāpūr.⁹⁹ This episode aligns well with the prevailing narrative in the historiography of medieval and early modern Zoroastrianism, portraying Zoroastrians as oppressed, marginalized, and sometimes enslaved communities. While acknowledging the intermittent adversities faced by Zoroastrians under Muslim rule, it is important to recognize that these challenges did not summarize the entirety of Zoroastrian life during this period. As an initial step towards providing nuance to this narrative, this paper has highlighted moments when Zoroastrians themselves engaged in discussions about their own slaves.

It has been argued that the mentions of slaves in the Persian Rivāyats go beyond being mere reflections of scholastic treatment of an obsolete legal issue. Evidence from Parsi questions spanning several centuries suggests that some affluent Parsis owned non-Parsi domestic slaves, predating their interactions with Europeans. The issue of converting these slaves to Zoroastrianism, a topic fraught with complexity for modern Parsis, constituted a central question concerning them.¹⁰⁰ It can be imagined that Parsis reacted in various ways to this problem. This variety of reactions is further intensified by the absence of a central authority among late medieval and early modern Parsis. Some masters may have preferred to convert their slaves. Among the Parsis who chose this path, a good number might have viewed their conversion as a practical solution to address the perceived impurities that could arise from the presence of non-Zoroastrian domestic slaves in their households. On the other hand, some others rejected outright the idea of their conversion, possibly due to the non-Zoroastrian background of the slaves (or their low social status).

In scholarly discussions, Parsi houses have been depicted as bastions of Zoroastrian purity before the seventeenth century, free from non-Zoroastrian elements and individuals. In light of the evidence presented, it has become clear that this assumption requires revision. The early exchanges with Iranians reveal that Parsi households were not devoid of non-Parsi individuals and elements. In addition to slaves, some passages in the Persian Rivāyats suggest that Parsis employed non-Zoroastrian wetnurses, utilized leather produced by joddīns, and used ghee processed by “infidels” in the preparation of their ritual and daily meals.¹⁰¹

The responses given by Iranian priests are often regarded as “academic”,¹⁰² and distant from the harsh realities of a marginalized and impoverished community in Persian society.¹⁰³ They themselves lived in conditions similar to slavery, making them far from being slave owners. While the academic nature of the priestly answers is undeniable, the present author does not entirely subscribe to the idea that all Iranian Zoroastrians were universally destitute and isolated. As previously mentioned, the sources occasionally refer to certain Iranian Zoroastrians who wielded a decent degree of wealth and power.¹⁰⁴ Is it possible to imagine that these wealthy Iranian Zoroastrians, much like some Parsis, might have employed domestic slaves? Given the current state of research on pre-modern Zoroastrianism in Iran, this is indeed a very unconventional question to ask. However, a curious Persian poem attributed to a certain Bahram Mehraban, probably composed before or during the eighteenth century, may give us a new perspective on this subject. The poem recounts the arduous journey of a certain Ardeshīr, an Iranian Zoroastrian merchant, as he navigates the perils of the sea on his return from India to Iran. In his quest to overcome these challenges, Ardeshīr turns to Bahram Izad, the revered Zoroastrian divinity known for safeguarding travelling Zoroastrians and protecting those facing adversity. The central theme of the story is the pivotal role played by Bahram in helping Zoroastrians.¹⁰⁵ Of great interest to our subject is the beginning of the story, where it is said that Ardeshīr bought several male and female slaves (gholāmān va kanizān) from India. He subsequently undertook their transportation, along with other merchandise, to Iran.