Slavery Represented in Bactrian Documents

ABSTRACT

This article addresses the idea and practice of slavery reflected in Bactrian documents. Bactrian documents are found in various parts of modern-day northern Afghanistan, a region broadly corresponding to late antique Bactria. The article discusses the definition of slavery and the slave's social status reflected in these documents. It shows that slavery was a practice that was authorized within the Bactrian legal system, where two types of slavery were recognized. The first type was that of the slave who was born in the owner's possession (χοβο νιζαδαγο μαρηγο), while the second was that of the slave who was purchased for money (δδραχμο χιρσιγο). In both cases, the slaves were regarded as personal property, with their owners having total authority over them. The Bactrian legal system allowed a slave to purchase his freedom by paying his full price and also permitted families to sell their children into slavery in times of difficulty. The legal documents were produced in the court of law, where some officials witnessed and sealed them, providing evidence that slavery was widely practised in late antique Bactria.

KEYWORDS:

• Slave
• slavery
• Bactrian documents
• Bactrian legal system
• late antique Bactria

Introduction

The idea and practice of slavery in late-antique Bactria is a subject that has received little scholarly attention.¹ The prime obstacle to an in-depth study of this subject was the scarcity of primary sources. However, since the 1990s, a number of Bactrian documents found in different parts of modern-day northern Afghanistan have provided fresh information on the practice of slavery in late-antique Bactria. The present article attempts to enhance our understanding of slavery and its relevant regulations as represented in these documents. In doing so, it specifically asks the following questions. How do these documents define slavery? What kind of terminology is used to refer to slaves? How did a person become a slave? And what was the legal status of a slave?

To provide some answers to these questions, this article will discuss these issues in two parts. In the first part, it will analyse two Bactrian documents that are directly related to an individual’s enslaved status, the first of them pertaining to an adult slave who purchased his freedom, and the second to a young boy who was sold into slavery. In the second part, the article will contextualize these two documents by using other relevant Bactrian documents to situate the practice of slavery within its social context in Bactria. Through a close reading of these documents and a comparison between them, I seek to demonstrate how slavery operated in each set of circumstances and the similarities and differences between those circumstances in late-antique Bactria.

The sources consulted in this article are several Bactrian documents. Bactrian is the only Middle Iranian language that was written in the Greek alphabet. Bactrian documents are found in different parts of modern-day northern Afghanistan, a region that broadly corresponds to late antique Bactria.² In its widest sense, Bactria was a region stretching from the Marw al-Rūd river in the west to Badakhshan in the east, the Hindukush to the south, and the Hisar mountains to the north, with the Amu Darya flowing through the middle of it.³ The documents in question are dated between the early fourth and the late eighth century CE – a period during which the Sasanians, the Kidarites, the Hephthalites, the Western Turks and the Arab Muslims attempted to control this region. In terms of subject matter, the documents vary from administrative, economic, and legal documents to official and private letters reflecting the diverse socio-political circumstances in Bactria.⁴ Bactrian documents have been translated into English by Nicholas Sims-Williams, and several scholars have already addressed the socio-political and economic elements reflected in them.⁵ While this article has benefited from the existing studies on Bactrian documents, it explores these materials specifically with a view to understanding the idea and practice of slavery in late-antique Bactria.

Bactrian Documents Related to Slavery

The first document to discuss is a deed of manumission. It was written in the year 247 of the Bactrian calendar (470 CE), in the city of Lan in Kadagstan, at the court of the governor.⁶ At this time, Bactria was under the control of the Hephthalites, who defeated the Sasanians.⁷ The document is a double contract witnessed by a number of officials including the overseer of the market (οασαροβιδο).⁸ In a double contract, the same text is written twice on the same parchment. The upper part (called the closed part) is rolled up and sealed, while the lower part (called the open part) is left open for reading. The closed part would only be opened before a judge in the case of a dispute.⁹ The double contract reduced the possibility of forgery as the same text was written twice in the same language and was handwritten. Each party received a copy and the record of the document was kept in the governmental archive for future consultation.¹⁰

This deed of manumission states that Asbid Nunukan granted freedom to a certain slave named Zer, son of Bag-bandag, who had paid an amount of money to buy his freedom. The document acknowledges the names of Zer’s former owners who were natives of Pidud, an area that is known from another Bactrian document and probably was located in the northwest of Balkh.¹¹ It is not known how and when Asbid Nunukan purchased Zer from his former owners and how Zer ended up in Lan in eastern Bactria. Most probably, Asbid Nunukan had a document related to the purchase of Zer from his former owners because the document gives details about these owners.¹²

The document contains specific legal statements and clauses that elaborate further on the regulations related to the practice of slavery. The document explains that Zer was a slave (μαρηγο) who purchased his freedom permanently (μαβαροζαμανο). As a result, he became a free person (αζαδο). How he acquired the money to achieve this is not mentioned. Nor is it known what kind of work Zer had performed at Abid Nunukan’s house. Asbid Nunukan, the owner of Zer, declares that he has made this document freely and willingly and is satisfied with its contents. He says that because Zer paid his full price, he has set him free forever without having any claim and argument. He promises that if he or his family members or relatives pursue Zer in the future, object to his freedom, or claim him by showing another sealed document, all such claims will be invalid at the court of law and he should pay a fine of 50 gold dinars to the government and the same amount to the opposite party – which is to say, to Zer.¹³ In Bactrian documents, the amount of the fine was often twice that of the price paid for a deal. It is reasonable to infer, therefore, that Zer had paid 25 dinars for his freedom.

The document does not mention what kind of social rights Zer could gain after he had bought his freedom. It is clear that Zer’s social status had changed and he was recognized as a free person by the local authorities. The document ensured that Zer should not be pursued or claimed by anyone, indicating that this document also played the role of a protection document issued by the government.¹⁴ In terms of legal rights, Zer purchased his freedom and should have been given a copy of the document presenting his freedom. However, the document is not sealed by Zer, which demonstrates the fact that although he was recognized as a party in the court of law, he still did not have the right to seal a legal contract. He may have obtained that right later if he entered into a legal deal.

The second document to discuss was produced two centuries later and was written in 446 of the Bactrian calendar (669 CE) in the district of Samingan, at Marogan in the market belonging to the rulers of Rob (ρωβοχαραγγο). The translation of this document reads as follows:

(It was) the year 446, the month Ab, the day Wahman, [when this sealed document, this purchase contract, was written] here in the district of Samingan, at Marogan, [the market of the khars of Rob, with the cognisance of the god] Ram-set, the granter of favours (and) granter of wishes, the wonderful, whose [worship] is [here at Maro]gan, at the market, with the cognisance of Zhun-lad Shaburan, [the tapaghlïgh] iltäbir of the [qaghan] prosperous [in glory], khar of Rob, and in the presence of Khusaru the tarkhan, and in the presence of D[ebraz … ,] senior overseer of the market of the khars [of Rob], and also in the presence of the other freemen who were present [there amongst them] and (who) bear witness concerning this matter.

Now: there has now been sold by me, Y[askul, and] by me, Yezd-gird, the sons of Kaw, inhabitants of Khwastu, who are now present here in [the district of Samingan,] and our brothers (and) sons, to you, Fanz, and to [you, Wind]-marg, and to you, Pusk, the sons of Bag-mareg, you whose house they call Gaba[liyan], and to your brothers, sons (and) descendants a certain boy [belonging] to us as brothers, this (same) boy who is called Khalas (since) we have not been able to keep him in plenty (and) famine – for three Persian dirhams.

So now may the boy described herein belong properly and well to you, Fanz, and to you, Wind-marg, with (your) brothers, sons, and descendants thereafter, from now to eternity. And also you have the right (to do) whatever it may suit you (sg.!) to do, to keep (him) yourself, to sell (him), to pawn (him), to give (him as) a gift, to put (him for) purchase (or for) hire, to detain (him) for a misdeed, (or) to let (him) free (in return) for service. Then you (sg.) have the right to do to him in every way what it is the custom (to do) to (slaves) purchased for money and to slaves born (in) one’s own (possession).

And now, if there should be anyone, from tomorrow or at (any) future time, who might dispute concerning Khalas with you, Fanz, [or with] any of the members of your household, (or if) they [might] get hold of (him) by theft (or) by force (?), [then I,] Yaskul, and I, Yezd-gird, and the members of our household [shall cause him to be] released [and detached from the claimant; and then], if we are not able to cause (him to be) released (and) detached and [we are not able to] do whatever [we agreed to do, then we shall pay] to you, Fanz, [and] to the [members of ] your (pl.) [household …] [… this purchase] contract … 

Verso

Yaskul (seal A)

The god Ram-set (seal B)

The khar of Rob (seal C)

[Khusa]ru the tarkhan (seal D)

[Deb]-raz (seal E)¹⁵

The setting of this document is the marketplace that stood under the jurisdiction of the rulers of Rob. In this document, we read about the selling of a boy named Khalas. The boy in question is sold into slavery by his family members due to the famine (οιγνο) that has affected their area called Khwastu in eastern Bactria.¹⁶ They have travelled from Khwastu to Marogan market in Samingan where the boy has been sold. At this time, Marogan was a well-known market that not only had a place for selling slaves, but evidently also had moneylenders. A Bactrian loan contract produced three years after the document under discussion shows that a man and his wife travelled from Tarmidh (modern Termez), an area in the north of the Amu Darya, to loan forty silver dirhams.¹⁷ The boy was sold to another family in the presence of the market’s overseer and a number of other witnesses.

The document relating to the sale of the boy Khalas is another double contract and its upper part was originally rolled and sealed by five or six clay sealings. Like the first document, the second document also bears some specific legal statements and clauses reflecting the practice of slavery in seventh-century Bactria. Before discussing these regulations, we need to have a closer look at the structure of this document to understand the circumstances in which this document was made.

The document has the standard layout, structure, and clauses of other economic Bactrian documents. It is a ‘purchase contract’ (μαχιρσο βωσιτγο) and contains conventional phrases and formulae that are common in other Bactrian documents related to purchase of properties such as land. Such verbal conventions give the impression that the boy was sold as property. The fact that just three dirhams were paid for the boy indicates that he must have been very young at the time of the sale. Like other economic contracts, this contract is permanent, and breaking its conditions would have entailed legal consequences such as the payment of a fine to the government and the opposite party.

The opening part of this document reflects the legality of selling children into slavery in a time of crisis. The document opens with recognition of the god Ram-set. It then refers to Zhun-lad Shaburan, the ruler of Rob who recognized the overlordship of the Turkic qaghān. Subsequently, it mentions the names of Khusrau the tarkhān or military commander, and Deb-raz the senior overseer of the market. It refers to witnesses without naming them. However, the word azādkār (αζαδοκαρο), meaning ‘freeman’, shows that the witnesses were freemen who had the right to witness a legal deal. The names given in this part of the document match the names of the people who sealed this document. Evidently, therefore, a family selling their son into slavery was a legal act recognized by the officials, the priest of the god Ram-set and regular people who were in the market.

The second part of the document helps us understand the relationship between the boy and his sellers. It refers to Yaskul and his brother Yezd-gird who brought the boy for sale in the market. Though the existing translation of this document suggests that the boy belonged to them as brothers (χοβομηνο βραδδιγο), some scholars believe that the boy was not their brother but a slave when he was sold in the market.¹⁸ A close look at the terminology used to refer to the boy indicates that the boy was the sellers’ brother.¹⁹

First, if the boy was a slave, then one might have expected the document to refer to him as māreg (μαρηγο) or bandag (βανδαγο). These two specific terms are often used for slave/servant in other Bactrian documents.²⁰ Second, the suffix -ιγο added to brād (βραδδ) is noteworthy. The word βραδδιγο is a diminutive form for βραδγο, and it can be translated as ‘little brother’. Thus, the Bactrian sentence χοβομηνο βραδδιγο ιϵο βαλακο ασιδο χαλασο ναμο can be translated as ‘a boy who is called Khalas is our little brother’.²¹ Third, the justification clause confirms our impression that Khalas was the vendors' brother. It mentions that the vendors sold the boy because they could not keep him in times of famine. If the boy was not a family member, then the document would not have used this justification clause.²² All this helps us to conclude that Yaskul and Yezd-gird consciously sold their little brother Khalas into slavery in the market because they could not keep him anymore. This part of the document reflects the fact that selling a family member in a time of crisis was acceptable to the officials, the priest, and other people who witnessed the document. In other words, there was no legal objection to such a course of action.

The document relating to the sale of Khalas had legal significance. It transferred the right of guardianship from the original family to another family. However, the nature of that right was largely changed by the transaction. As a result of this agreement, the freeborn boy became a slave and the property of another family, and his new owners obtained rights to treat him in any way they wished. They could keep the boy in their house for domestic service, sell him, pawn him, give him as a gift, hire him out for money, detain him for his misdeeds, or free him for good service.²³

It is worth mentioning that the sale of children into slavery in times of difficulty was not confined to Bactria. The Middle Persian legal text Mādagān ī Hazār Dādestān (‘The Book of One Thousand Judgments’) has a chapter on selling children. In one episode the condition for selling children is described as follows:

Only the father is entitled to sell children into slavery. And, on his deathbed or in a state of physical helplessness, he is entitled to sell them only to someone to whom it is permissible to sell in the state of atvādāt.²⁴

This passage clarifies several issues. It limits the right of selling children to the father alone, and furthermore only under certain conditions. If a father is destitute and cannot feed the children, then he can sell them. However, he is not allowed to take a child to the market and instead is permitted to sell the child only to his kinsman or to the kinsman of the child’s mother. In contrast to this Middle Persian passage, the Bactrian document related to the sale of the boy does not clarify the relationship between the sellers and the boy’s brothers. Likewise, it does not speak about the boy’s father. It is not known if the boy’s father was alive when Khalas was sold. It is also unknown how the elder brothers acquired the right to sell their little brother.

The document related to the sale of the boy refers to two types of slavery that were legal and practised in seventh-century Bactria. The first type was that of the slave who was born in the owner’s possession (χοβο νιζαδαγο μαρηγο), while the second was that of the slave who was purchased for money (δδραχμο χιρσιγο). In other words, there were slaves who were born into slavery, and there were also slaves who were born free but sold into slavery later. The boy Khalas belonged to the second category. The document does not clarify what kind of rights were possessed by slaves belonging to these two categories. It does not say anything about food, clothes, or the place where the boy would live. The absence of any clauses relating to these rights suggests that there were no fixed regulations in this regard and the slave owners accommodated the slaves however they wanted to. The documents relating to Zer and Khalas show that slavery was a legal act commonly practised in late-antique Bactria. Combining these two documents with other relevant Bactrian documents helps us understand the social status of a slave in Bactrian society.

Slaves in Bactrian society

By putting these documents together and comparing them with other relevant Bactrian documents, we can learn about the nature of slavery presented in these documents. In both these documents and other Bactrian documents, the terminology to denote a slave is māreg (μαρηγο). This term derived from Old Persian mārikā, which means slave/servant.²⁵ However, māreg is frequently used in the greeting part of the Bactrian administrative letters, meaning ‘servant’.²⁶ In such cases, it appears that the term did not mean ‘slave’ in our current understanding, but rather that it was used by lower officials to address a higher official showing his obedience and respect. The term māreg is also used in combination with the names of deities. The document relating to the sale of the boy refers to Wind-māreg (Slave of the Wind) son of Bag-māreg (Slave of the God) who was one of the buyers of the boy. Personal names like these did not indicate any legal status but were used as an indicator of religious devotion.²⁷

The term bandag (βανδαγο) is also used for a servant in several Bactrian documents. It derived from Old Persian bandaka, meaning slave/servant.²⁸ All twenty-nine wooden receipts dated to ca. 470 in Kadagstan, an area in eastern Bactria, note that goods were delivered by different bandags, without individually naming these individuals; by contrast, the bandags’ masters are identified by name in every case.²⁹ Like māreg, the term bandag can be seen in Bactrian personal names such as Bag-bandag (Slave of the God) and Zhun-bandag (Slave of the god Zhun).³⁰ A number of other Bactrian documents refer to female slaves as banz (βανζο) without further elaborating on it.³¹ In addition to such nomenclature, the term marskond (μαρσκονδο) is also used to refer to slave, servant, people and staff.³² But this term is not as common as the māreg and bandag, which are general terms referring to slave and servant alike.

In contrast to māreg, the word āzād is used in these documents to denote the sense of ‘freeborn’. The frequent usage of māreg in juxtaposition to āzād in the Bactrian documents show that these legal statuses mattered: people were recognized as either freeborn or slave in the Bactrian legal system.³³ It is likely that freeborn and slave were two social labels that rendered these categories visible and meaningful in Bactria. A freeborn and a slave did not have equal social rights. Their social statuses defined their position in the society.

Bactrian documents show that the freeborn was often identified by association with a certain household. The household was the main social unit and the social relations were largely understood around the axis of the household. The Bactrian terms kadag (καδαγο) and khān (χανο) are frequently used for the household in the Bactrian documents.³⁴ The suffix -ιγανο could be added to kadag (kαδαγιγανο) or khān (χανιγανο), thereby conveying the meaning ‘members of the household’.³⁵ The household was a space where the family members like grandparents, parents, and children lived together and where household provisions were kept. The household was an element of social identity as well. The individuals were identified with their association to certain households.³⁶ Each household had a specific name, often called after a father or grandfather.³⁷ Thus, association with a household was important in the society.

A freeborn person could marry and have children.³⁸ He could own property including slaves by inheriting, purchasing, or receiving gifts and donations.³⁹ A freeborn had the right of ownership over his properties. If he inherited land then it was recognized as ‘ancestral land’ (πιδοραστο).⁴⁰ It also was called ‘ancestral estate’ (βοναγο πιδοριƥτο) in Bactrian documents.⁴¹ He had all rights on his properties and these rights were recognized in the Bactrian legal system.⁴² A freeborn person could work on his land or lease a piece of land.⁴³ He could be involved in trade, work as a labourer or join the administration, or military.⁴⁴ The freemen were called citizens (ƥαριγανο) and they had certain legal rights including to act as a witness to a legal agreement.⁴⁵ A freeborn person was not treated in the same way as a slave. A Bactrian legal document produced in 503 of the Bactrian calendar (726 CE), in Khesh, an area between Bamiyan and Kabul, clarifies that the freeborn does not perform the same duties assigned to a slave. The document states that a freeborn woman and her children had ‘free feet and free hand’ (αζαδο παλο οδο πιδο αζαδο λιστο), this verbal formulation conveying the extent of the freedom that they enjoyed in their houses.⁴⁶

Unlike the freeborn, the slave did not have all these social rights. Bactrian documents do not identify a slave with any household. Though a male slave could have his own family and perhaps his personal belongings, he still counted as somebody else’s personal property. The documents indicate that the slaveowner had all rights to treat his slaves the way he wished. The document relating to Khalas clarifies that his new owners could keep him at home for domestic work, sell him to another person, pawn him, give him as a gift, put him for purchase in the market, or set him free in return for service.⁴⁷

Female slaves were treated the same way. A Bactrian legal document produced in 478 of the Bactrian calendar (700 CE) in Kadagstan shows that Bagazyas, the Turkic queen of Kadagstan, gifted a woman (ζινο) to a priest named Kamird-far after the queen’s infant was healed by the priest. The document mentions that the priest could keep this woman as a slave-girl (βινζο) for his pleasure. He had the right to keep her at his house, sell her, gift her, or set her free in return for service.⁴⁸ The slaveowner also was allowed to detain and beat her for misdeeds.⁴⁹

Slaveowners could discharge liabilities or debts by giving away their slaves. At least four Bactrian legal documents attest to this practice. The first such document is a loan contract produced at Marogan market in Samingan in 449 of the Bactrian calendar (671 CE). The contract authorizes the loan-givers to fix a non-refundable penalty on the debtors if they could not pay their dues on time. The penalty included cattle (στωρο) and people (μαρσκονδο) belonging to the debtors. Th reference to people here is evidently to slaves and the fact that the word cattle is followed by people supports this idea. The document uses abandarnīg (αβανδαρονιγο) ‘outside’ and binīg (βηνιγο) ‘inside’ when referring to cattle and people respectively which may imply that the cattle were kept outside, and the slaves lived inside the house.⁵⁰

The second document shows a similar situation. It is a lease contract produced in 490 of the Bactrian calendar (712–13 CE), in Rob. At this time, the Umayyads were expanding their authority to most parts of Bactria. The contract states that a vineyard was leased by a family on certain conditions. The leasers were obliged to pay the full price in kind, with the price reckoned as one person (μαρσκονδο), one plough-ox (γαο κιƥαγο) and six sheep (ποσο). They could grow vines and keep one-third of the yield and deliver the rest to the landowner.⁵¹ It is important to note that this document has slaves, ox, and sheep in the same line indicating people’s understanding of slaves as personal property that could be handed over in discharge of liability.

The third document provides further details and was produced in 500 of the Bactrian calendar (722 CE) in Guzgan in western Bactria. The authors of this document declared that they had a dispute over a debt with a certain family. After visiting the local authorities, they received a fine from that family and agreed to produce this document. The declarers mention that they will not have any dispute with that family in the future. In particular, they state that they will not issue claims against the other party regarding slaves. This verbal commitment is repeated three times indicating that the dispute over the debt must have been related to slaves. Possibly, they promised not to claim slaves for the debt anymore because they had already received a fine from that family. The document reveals that slaves were counted as important property and their owners tried to hold onto them.

The final document supports this latter inference. It is a protection letter issued by the Turkic king of Kadagstan in 549 of the Bactrian calendar (772 CE). At this time, the Abbasids already controlled most parts of Bactria except Kadagstan. The document explains that a certain Mir son of Bek presented the king with a petition relating to a family dispute over properties. Mir’s brother Bab had some debts and his creditors were seeking redress. Mir requested the king to protect him, his family, his people, and his properties. The term used for Mir’s people is marskond which is different from the term used for Mir’s family (μιραγγο βραδο πορανο). We have already seen in some of the other documents that the term marskond could be used to mean a slave. Mir’s usage of this term suggests that Mir’s people were his slaves and that Mir did not want people to take his slaves for the debt of his brother Bab. Similarly, he wanted the king’s protection for his slaves.

On account of their new legal status, Zer and Khalas may have had similar experiences. Zer could enjoy the freedom guaranteed by the officials. His new status was announced not simply orally but in the form of a written document issued in the court of law. This suggests that the freedom of a slave had to be acknowledged in the court of law where a legal document was produced and given to the slave as evidence of his freedom. Similarly, Khalas’ slave status was decided by his elder brothers and simultaneously confirmed by the court of law.

The documents relating to the sale of the boy Khalas and the manumission of Zer are legal contracts produced in the government offices. The contract relating to the boy refers to the god Ram-set as a witness, which creates a religious obligation for abiding by the agreement. In both cases, the overseers of the market have presented a procedure that part of the process involves selling slaves. Not only were the overseers aware of the market prices and the regulations of slavery, but their presence also meant that they recognized the freedom of Zer and the slave status of Khalas. The presence of the overseers would help prevent people claiming Zer as a slave later. Similarly, people could no longer claim that the boy was a freeman, because the overseers would have had a network of associates throughout the region who could attest to the boy’s newly enslaved status.

Conclusion

The documents discussed in this article show that slavery was an accepted and legal practice in Bactria. Despite regime changes, slavery continued in the region throughout the period of late antiquity. Slaves were identified with specific terminologies such as māreg, bandag, banz and marskond. The terms māreg and bandag were combined with deities’ names to make personal names. The term māreg was frequently used by lower officials to address a higher official in the greeting part of the Bactrian administrative letters showing his obedience and respect.

Slaves were not regarded as citizens. They were not identified with their own family but rather were recognized by their association with their owners. Nor did they have the right to witness and seal a legal document. The Bactrian legal system preserved all rights for the slaveowners to treat their slaves as they wished. They also could pay their liabilities by giving away their slaves and they could beat or detain their slaves for their misdeeds. However, they had to protect their slaves against external claims and disputes because slaves were their property.

The documents relating to the sale of the boy Khalas and the deed of manumission relating to Zer reflect the transformation of the social status of two individuals within the Bactrian legal system. The document relating to Zer shows that a slave could become free forever if he paid his price. It indicates that freedom was purchasable and that a slave was not an absolute and fixed-value commodity, but rather a human who could change his social status. However, it is not clear what kind of social rights he could acquire thereafter. Was he treated equally as a freeborn or was he humiliated because of his slave background?

In contrast to Zer, the document relating to the sale of the boy Khalas shows that a freeborn could become a slave. This document demonstrates that the Bactrian legal system allowed the transformation of the right of guardianship of the boy to another family not as a family member, but as a slave. With this document, the boy lost all the social rights of a freeborn that he could have had in his own family had there not been a famine. Much remains unknown about the fate and status of Zer and Khalas. For instance, what could have happened to the boy after his arrival at his new house, or what could Zer do after he left the court of the governor in his newly bought state of freedom?

These issues urge us to think not only about the descriptive aspect of slavery reflected in these documents, but also about the psycho-sociological impact of slavery. For example, how did a slave live after being emancipated? And how did a freeperson continue his life after being enslaved? Perhaps future research will provide some answers to these questions.

Acknowledgements

I would like to thank Nicholas Sims-Williams for his valuable comments. I am also grateful to the reviewers of this Special Issue and the journal’s editors for their suggestions and comments.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research was supported by the European Research Council under grant number 683194.

Notes on contributors

Said Reza Huseini

Said Reza Huseini is a PhD candidate at Leiden University and a research fellow at King’s College, Cambridge, King’s College, King’s Parade, Cambridge CB2 1ST, UK Email: [email protected]

Notes

1 Some scholars have already looked at Bactrian documents to understand the socio-political and economic situation in Bactria and have briefly referred to slavery in their work. See, for instance, Nicholas Sims-Williams, ‘A Bactrian Deed of Manumission’, Silk Road Art and Archaeology 5 (1997/98): 191–211; Khodādād Rezākhāni, ‘The Bactrian Collection: An Important Source for Sasanian Economic History’, Encyclopedia Sasanika 13 (2018): 1–14; Arezou Azad, ‘Living happily ever after: fraternal polyandry, taxes and “the house” in early Islamic Bactria’, BSOAS 79, no. 1 (2016): 33–56; Nicholas Sims-Williams and Étienne de la Vaissière, ‘A Bactrian Document from Southern Afghanistan’, Bulletin of Asia Studies 25 (2011/2015): 39–53.

2 Nicholas Sims-Williams, ‘New Documents in Ancient Bactrian Reveal Afghanistan’s Past’, IIAS Newsletter 27 (2002): 12–13.

3 A useful description of Bactria in the early seventh century is reflected in the records of the Chinese Buddhist pilgrim Xuanzang who visited Bactria in 629–30 CE on his way to India. According to him, the Tu-ho-lo or the country of Tukharians (which is the region of Bactria) was 1000 in 3000 li. On the east, it was bounded by the T’sung-ling mountains and on the west, it touched Po-li-sse or Persia. The great ‘Snowy Mountains’ was located to the south and to the north the ‘Iron Gates’ were located. The Amu Darya flew through the middle of this country. SI-YU-KI, Buddhist Records of the Western World, trans. Samuel Beal (London: Kegan Paul, Trench, Trübner, 1906), 37.

4 Our extant holding of Bactrian documents do not cover the whole region. Their contents show that they were produced in Guzgan (modern Juzjan) and Gaz (possibly Darray-i Gaz, south of Balkh) in the west, Rob region (modern Samangan) in the south, Kadagstan (possibly in modern Baghlan), and Warnu (probably Qala-i Zal or Qunduz) in the east of Bactria. So far, no Bactrian document has been found from the Balkh oasis. Bactrian documents were translated and published by Nicholas Sims-Williams in several publications. See Nicholas Sims-Williams, ‘Nouveaux documents bactriens du Guzgan (New Bactrian Documents from Guzgan)’, Comptes rendus de l’Académie des Inscriptions & Belles-Lettres (2002): 1047–58; Nicholas Sims-Williams, ‘Bactrian Letters from the Sasanian and Hephthalite periods’, in Proceeding of the 5th Conference of the Societas Iranologica Europa, Ravenna, 6–22 October 2003, vol. 1:. Ancient and Middle Iranian Studies, ed. Antonio Panaino and Andrea Piras (Milan: Mimesis, 2007), 701–13; Nicholas Sims-Williams, Bactrian Documents from Northern Afghanistan, vol. 1: Legal and Economic Documents, rev. ed. (Oxford: The Nour Foundation, Azimuth Edition, Oxford University Press, 2012), henceforth DBI; Nicholas Sims-Williams, Bactrian Documents from Norther Afghanistan, vol. 2: Letters and Buddhist Texts (London: The Nour Foundation with Azimuth Editions, 2007), henceforth DBII; Nicholas Sims-Williams and François de Blois, Studies in the Chronology of the Bactrian Documents from Northern Afghanistan, with contributions by Harry Falk and Dieter Weber (Vienna: Verlag der Österreichischen Akademie der Wissenschaften, 2018).

5 Geoffrey Khan, ‘The Pre-Islamic Background of Muslim Legal Formularies’, ARAM 6 (1994): 193–224; Rezākhāni, ‘The Bactrian Collection’, 1–14; Khodadad Rezakhani, ‘Balkh and the Sasanians: Economy and Society of Northern Afghanistan as Reflected in the Bactrian Economic Documents’, in Ancient and Middle Iranian Studies, ed. M. Macuch, D. Weber, and D. Durkin-Meisterernst (Wiesbaden: Harrassowitz, 2010), 1–21; Kevin Van Bladel, ‘The Bactrian Background of the Barmakids’, in Islam and Tibet: Interactions along the Musk Routes, ed. A. Akasoy, C. Burnett, and R. Yoeli-Tlalim (Burlington, VT: Ashgate, 2011), 43–88; Richard Payne, ‘The Making of Turan: The Fall and Transformation of the Iranian East in Late Antiquity’, Journal of Late Antiquity 9, no. 1 (2016): 4–41; Azad, ‘Living Happily Ever After’, 33–56. Scholarly attention to these documents has recently increased. For instance, see Khodadad Rezakhani, ReOrienting the Sasanians: East Iran in Late Antiquity (Edinburgh: Edinburgh University Press, 2017); Arezou Azad, ‘The Beginning of Islam in Afghanistan: Conquest, Acculturation and Islamization’, in Afghanistan’s Islam from Conversion to the Taliban, ed. Nile Green (Oakland: University of California Press, 2017), 41–55; Hossein Sheikh, ‘Studies on the Bactrian Legal Documents’ (PhD diss., Georg-August-Universität Göttingen, 2017); Said Reza Huseini, ‘Acts of Protection Represented in Bactrian Documents’, Annales islamologiques 54 (2021): 107–24; Said Reza Huseini, ‘The Idea and Practice of Justice Represented in Bactrian Documents’, Association for Iranian Studies Newsletter 41, no. 2 (2020): 28–31; Said Reza Huseini, ‘Between Turks and Arabs: Household, Conversion and Power Dynamics in Early Abbasid Bactria (700–772)’, in The Ties that Bind: Mechanisms of Social Dependency in the Early Islamic Empire, ed. Edmund Hayes and Petra M. Sijpesteijn (Cambridge: Cambridge University Press, forthcoming in 2023); Hossein Sheikh, Studies of Bactrian Legal Documents (Leiden: Brill, 2023).

6 Bactrian Document no. F in Sims-Williams, BDI, 38–41. This document is now kept at the Hirayama Ikou Silk Road Museum, no. 103696.

7 Khodadad Rezakhani, ‘From the Kushans to the Western Turks’, in King of the Seven Climes: A History of the Ancient Iranian World 3000 BCE–561 CE, ed. Touraj Daryaee (California: UCI Jordan Centre for Persian Studies, 2017), 206; Robert Huag, The Eastern Frontier: Limits of Empire in Late Antique and Early Medieval Central Asia (London: I. B. Taurus, 2019), 60–61.

8 For a physical description of this document, see Nicholas Sims-Williams, ‘A Bactrian Deed of Manumission’, 191–211. Sims-Williams dated this document to 479 in this article, but he also suggested 470 for this document. For recent studies on Bactrian chronology, see Sims-Williams and de Blois, Studies in the Chronology of the Bactrian Documents, 21–42.

9 For the layout and structure of Bactrian documents, see Sims-Williams, BDII, 15–17; Sims-Williams, ‘Bactrian Letters from the Sasanian and the Hephthalite Periods’, 701–713. For details on the making of a double contract, see https://www.trismegistos.org/seals/overview_2d.html (accessed online 24 February 2022).

10 Huseini, ‘Acts of Protection’, 107–124. See also Also see Huseini, ‘The Idea and Practice of Justice’, 28–31.

11 Bactrian Document no. O in Sims-Williams, BDI, 80–83.

12 Bactrian Document no. F in Sims-Williams, BDI, 38–41.

13 Bactrian Document no. F in Sims-Williams, BDI, 38–41.

14 For protection, see Huseini, ‘Acts of Protection’, 107–124.

15 The English translation is by Nicholas Sims-Williams. See Bactrian Document no. P in Sims-Williams, BDI, 84–87.

16 There is a district called Khwust in present-day Baghlan province that can be identified with Khwastu mentioned in the Bactrian document.

17 Bactrian Document no. Q in Sims-Williams, BDI, 88–91; Bactrian Tarmidh corresponds with modern Tirmidh.

18 I suggest that Khalas was a young boy sold by his elder brothers, but my argument has been questioned by some scholars who have argued that the boy was already a slave and sold by his owners. Online Workshop, ‘Textual Sources and Geographies of Slavery in the Early Islamic Empire, ca. 600–1000’, Leiden University, 3–4 December, 2020.

19 Sims-Williams, BDII, 201.

20 Almost all Bactrian letters used these terms. Sims-Williams, BDII.

21 I am grateful to Nicholas Sims-Williams who kindly explained this sentence from a linguistic perspective. His explanation supports the idea that the boy was the sellers’ younger brother. Personal communication, 16 February 2022.

22 Sims-Williams raised this point (personal communication, 16 February 2022).

23 Bactrian Document no. P in Sims-Williams, BDI, 84–87.

24 Anahit Perikhanian, The Book of A Thousand Judgments (A Sasanian Law-Book), trans. Nina Garsoïan (California: Mazda Publisher, 1997), 94–95. I would like to thank Nazanin Tamari for informing me about the Middle Persian text contained in this volume that relates to the topic of this present article.

25 Sims-Williams, BDII, 231.

26 Almost all Bactrian letters have this term in their opening parts. Sims-Williams, BDII.

27 For Bactrian personal names, see Nicholas Sims-Williams, Bactrian Personal Names (Iranisches Personennamenbuch, vol. 2: Mitteliranische Personennamen, part 7) (Vienna: Verlag der Österreichischen Akademie der Wissenschaften, 2010).

28 Sims-Williams, BDII, 201.

29 Bactrian Documents no. am 1–38 in Sims-Williams, BDI, 168–69.

30 Sims-Williams, Bactrian Personal Names, 65–66.

31 Bactrian documents no. A, N, X in Sims-Williams, BDI, 26–29, 68–73, 136–41.

32 Bactrian documents no. Q, Y in Sims-Williams, BDI, 84–87, 142–43.

33 Almost all legal documents refer to both of them.

34 Bactrian documents no. A, F, J, L, M, N, P, Q, T, U, V, W, X, Y in Sims-Williams, BDI, 26–29, 38–41, 48–53, 58–65, 66–67, 68–73, 84–87, 84–87, 88–89, 106–11, 116–17, 136–35, 136–41, 142–43.

35 Bactrian Documents no. jf in Sims-Williams, BDII, 132–33.

36 Bactrian Documents no. A, N, O, Q, V, W, X in Sims-Williams, BDI, 26–29, 68–73, 80–83, 88–91, 116–17, 126–35, 136–41.

37 Bactrian Document no. Nn in Sims-Williams, BDI, 74–75.

38 Bactrian Document no. A in Sims-Williams, BDI, 26–29.

39 Bactrian Document no. T, X in Sims-Williams, BDI, 88–89, 136–41.

40 Bactrian Document no. eb in Sims-Williams, BDII, 110–11.

41 Bactrian Document no. J, L in Sims-Williams, BDI, 48–55, 58–65.

42 All Bactrian documents related to the purchase of land ensure the owners’ right on their land. Examples are Bactrian Documents J, V, W in Sims-Williams, BDI, 48–53, 116–17, 126–35.

43 Bactrian Document no. U in Sims-Williams, BDI, 106–111.

44 Bactrian Document no. bh, cd, cg in Sims-Williams, BDII, 66–67, 74–75, 80–81.

45 Bactrian Documents no. C, L, X, Y in Sims-Williams, BDI, 32–35, 58–65, 136–41, 142–43.

46 Sims-Williams and de La Vaissière, ‘A Bactrian Document from Southern Afghanistan’, 43.

47 Apparently, slaves were kept primarily for domestic work. The preserved Bactrian documents do not refer to any kind of agricultural duties or military works assigned to the slaves. However, the number of documents reflecting slaves’ duties are few, and we thus lack the evidentiary basis to suggest that slaves were used only for domestic purposes. The existing documents mention that the slaveowners could use their slaves in any way they wished and this may have extended to work outside houses. Using slaves for domestic work is noted by Rezakhani, ‘Bactrian Collection’, 8. It is also mentioned by Azad, ‘Living happily ever after’, 40.

48 Bactrian Document no. T in Sims-Williams, BDI, 88–89.

49 Documents no. P, T in Sims-Williams, BDI, 84–87, 88–89; Sims-Williams and de La Vaissière, ‘A Bactrian Document from Southern Afghanistan’, 39–53.

50 Bactrian Document no. Q in Sims-Williams, BDI, 88–91; On the basis of this document, Azad mentiones that slaves served: ‘as in-kind currency for example to pay a fine for defaulting on a loan’. Azad, ‘Living happily ever after’, 40.

51 Bactrian Documents no. U in Sims-Williams, BDI, 106–11.