Flight

Flight remains one of the most visible issues of slave-ownership, appearing in law codes and hagiography, though sociolegal structures in which even free people were sometimes unable to leave their lord’s land or service without permission complicate our discernment of real cases rather than idealized narratives or legal generalisations.Footnote ²⁰ Charters issued by William II in the last decades of the eleventh century refer to the discovery and return of individuals living on other estates without permission.Footnote ²¹ Most likely, however, these were free men who sought new lordship and better terms of service; they appear to have left the estates with their livestock in tow, which does not suggest the furtive movements of a runaway slave.Footnote ²² As a result, our interpretation of slave flight is limited to perceptions by owners of runaways and the attempts of royal administrative structures to clarify responsibility for flight in a compensation-based legal system. This top-down perspective nevertheless indicates great concern for the retrieval of runaways and the protection of slave-owners’ property rights. Even if flight occurred rarely, it could significantly impact local communities and the maintenance of law and social order.

Where fugitive slaves were also offenders, primary responsibility and punishment lay with the slave. The West Saxon law code promulgated by Ine between 688 and 726 and reissued by Alfred in the late ninth century stipulates hanging as the punishment for runaway penal slaves and denies any kind of compensation to the owner.Footnote ²³ This tells us plainly that while English penal slaves may have been considered more eligible for manumission than, for example, enslaved captives, lawmakers still anticipated escape attempts.Footnote ²⁴ It also indicates that owners could not make claims regarding those who had proven themselves repeat offenders. Having already committed a crime severe enough to warrant enslavement (such as theft), runaway penal slaves were considered untrustworthy and irredeemable, and therefore faced a punishment sure to act as a deterrent for their fellows. That slave flight was considered a form of theft in West Saxon law is made obvious by the phrasing, ‘Gif witeðeow Engliscmon hine forstalie’, literally, ‘If an English penal slave steals himself’.Footnote ²⁵

More frequently, law sought to outline third-party accountability for flight and took a preventative approach. Ine aimed to establish responsibility for unintentional or deliberate actions which could precipitate an unfree person’s flight, namely allowing a servant to access tools for escape. Heavy compensation payments were due to a lord by anyone who lent a sword, a spear, or a horse to an esne who then absconded.Footnote ²⁶ The payments increased according to the ease with which the item facilitated flight and thus the loss of property. A sword would have enabled violent confrontation, but only at close quarters where the esne could still be subdued, and thus the lender owed only a third of the servant’s value. A spear would have allowed for an esne to injure or kill from a distance to slow or prevent pursuit, so the lender owed half. A horse would have allowed for an immediate and fast escape, and so the lender owed the esne’s entire value. On one hand, this law could have served as a preventative measure which forced individuals to consider the unintended consequences of arming the unfree, such as injury to others or a loss of property. On the other, this law could punish those who purposefully aided slaves in escape attempts. This interpretation suggests that Ine and the elites of his court worried that the unfree could form and utilize support networks to gain access to these luxury items.

Tenth-century law likewise sought to pre-empt any loss of property and establish a sense of communal responsibility for retrieving runaways. VI Æthelstan, promulgated 925 × 939, also placed the onus for deterring flight upon the slave and slaveholding communities. Under this law code, recaptured runaways were to be stoned to death in an execution identical to that of slaves convicted of theft, further illustrating the legal concept of a runaway as a special type of thief. Those responsible for stoning in turn owed a collective payment to the fugitive’s owner for the loss of his property.Footnote ²⁷ This also mirrors the treatment of slaves convicted of theft, and it complements Æthelstan’s assertion in the same law that all, presumably free, members of the community were to pursue the fugitive and owe compensation to the slave-owner should the slave elude their grasp.

This use of ‘duty to pursue’ against runaway slaves proves significant. Royal law of the tenth century reflects a shift in legal ideology in which the duty to pursue thieves moved from an assumed and accepted practice to an obligation. Tom Lambert argues that the duty to pursue was not a requirement of royal law until the promulgation of III Edmund (939 × 946) and the Hundred Ordinance, but understanding a fugitive slave as a type of fugitive thief suggests that the development of a ‘duty to pursue’ as an obligation of a participatory and communal legal system could be dated even earlier to VI Æthelstan, and attributed directly to perceived slave resistance.Footnote ²⁸

The standardization of this communal responsibility also appears somewhat later in the legal text Forfang, a code generally concerned with the retrieval of stolen property which was possibly composed in the second half of the tenth century. This text offered a reward of fifteen pence for any man who retrieved a fugitive or a horse, regardless of the distance.Footnote ²⁹ Certainly the retrieval of a man included thieves and other lawbreakers, and the inclusion of valuable mobile property such as horses suggests that the human fugitives in question could also include runaway slaves. Practically speaking, financial incentive must have functioned in conjunction with royal law and substantial communal pressure to seek out runaways. While allowing runaways, thieves, and other offenders to remain at large threatened the communal peace (OE frið) of the realm, runaway slaves in particular challenged the social hierarchy and permitted a fluidity of status beyond the bounds of accepted legal and social procedure.Footnote ³⁰ This reflects the situation in Ine’s law, which indicates that being party to abscondment constituted criminal behaviour, even if the lending of a weapon amounted only to gross negligence. With the delineation of their own status at stake, many free people, especially the poor, must have felt a sense of gravity in hunting a fugitive slave, and the possibility of a monetary reward may have ensured that they left their own work for as long as necessary.

We are limited in our understanding of when and why slaves fled, as well as who these fugitive slaves were, again because of the top-down perspective of our surviving sources. That lawmakers worried about the potential for and implications of slave flight throughout the early Middle Ages indicates that slaves did attempt to seek freedom through escape. Even if we prioritize elite fears of escaping slaves over a reality of ‘resistance’, it is still possible to hypothesize when and where slaves were most able to flee. We can presume that confusion caused by warfare and political instability allowed for some slaves to make their escapes. Such a situation in the late Roman period led to several recorded instances of mass flight by slaves from Italian cities, and similar situations may have occurred during early medieval regime changes, such as the various territorial conquests by viking and West Saxon armies from the mid-ninth through the mid-tenth centuries. On a more localized and routine level, the deaths of slave-owners and ensuing periods of property dispute and testamentary settlement could also have precipitated escape by slaves, either individually or in groups, a situation hinted at by one of the aforementioned charters of William II which, in granting an estate to new lord, demanded the return of all the men who had taken a previous lord’s death as an opportunity to leave.Footnote ³¹

Runaways may have been characterized by a gender disparity. Women may have felt greater ties to family, especially children, than did their male counterparts, which might have prevented them from leaving, particularly for those women who bore children by their owners. Children of extramarital liaisons between masters and their female slaves were not necessarily considered ‘illegitimate’ in the early Middle Ages, and surviving examples suggest that these offspring could participate in and benefit from family relationships. In the later medieval Gulaþing and Frostaþing laws from Norway, the children of enslaved concubines could still become heirs at the discretion of their fathers, even though a concubine’s status remained dishonourable and subordinate to legal wives.Footnote ³² Though the texts were written in the mid-thirteenth century at the earliest, these particular laws do not reflect the church’s preferred stance on marriage and may very well represent much older custom. A similar potential for paternal acknowledgement may be evident in the laws of Ine and Alfred, as well as Old English vocabulary, though in the late ninth-century will of one Ealdorman Ælfred, his bequest of folkland, a type of land tenure possibly governed by customary inheritance law, to his son Æthelwold by a woman not Ælfred’s wife remained contingent upon the king’s approval.Footnote ³³ We do see a more routine side of family inclusion in the ninth-century Irish Vita prima Sanctae Brigitae for a child who was not an heir. Although Brigit was the product of a lord’s extramarital relationship with his slave, Brigit’s father and free half-brothers still concerned themselves with her welfare and marriage prospects.Footnote ³⁴ The gendered division of labour may also have meant male slaves had a better knowledge of their surroundings which facilitated escape. A number of tenth- and eleventh-century documents refer to unfree men whose duties in herding, hunting, and ploughing may have allowed them to roam the boundaries of their estates.Footnote ³⁵ While women may also have herded as part of dairying, their specific duties, where discernible, were primarily domestic, such as the processing of these dairy products, weaving and sewing.Footnote ³⁶

Enslaved captives had the greatest potential for flight, since a lack of local kinship ties and investment in their new socioeconomic hierarchy would have made flight a more viable option. Many captives who had been transported within the British Isles were sufficiently close to home that escaping to their native land remained plausible, and we see this in two Irish examples. St Patrick, having been captured in Britain in the fifth century by Irish raiders, served as a slave in Ireland for six years before finding the opportunity to flee, after which he was able to return to his family.Footnote ³⁷ Similarly, St Findán was taken near his home in Leinster in the ninth century by Northmen and was able to evade his viking captors in the Orkneys while en route to Scandinavia. Though he never returned home, this was a conscious choice and not an impossibility; the author of his vita insists that God rescued Findán from drowning after Findán promised to never return to Ireland, and to instead travel to Rome and become a monk.Footnote ³⁸

Later elites feared that fugitive slaves could find safe haven in neighbouring kingdoms and enemy territory, and the role of political borders is especially evident in legal documents relating to the Danelaw. The Treaty of Alfred and Guthrum, which negotiated relations between the English and the Danes possibly between 880 and 890, includes a declaration that neither side will harbour the other’s slaves or freemen.Footnote ³⁹ II Æthelred, a treaty agreed with Olaf Tryggvason and his army c. 994, promises the same while adding thieves and individuals escaping feuds.Footnote ⁴⁰ The need to codify these terms of agreement at various points indicates that enemy harbouring of fugitives remained a significant concern for elites. We certainly see these worries highlighted in Wulfstan, archbishop of York’s Sermo lupi ad Anglos, a sermon probably composed and revised multiple times between 1009 and 1014.Footnote ⁴¹ In it, Wulfstan lamented his own troubled times when English slaves could find freedom in the Danelaw, only to then join campaigns in which they might meet and kill their former owners.Footnote ⁴² Wulfstan makes clear that attempts to close political borders to fugitive slaves were not necessarily successful, especially in the late tenth and early eleventh centuries, with such a situation leading to deeply disturbing issues such as fluidity of status, the overturning of social order, and the use of property against its ‘rightful’ owner.

Despite evidence suggestive of slave flight, authors did not express concerns regarding anything resembling a maroon community. Though such peripheral settlements of fugitive slaves appear in the large-scale slave systems of the ancient world and early modern period, they are not a visible feature of early medieval England.Footnote ⁴³ Their absence from the source material may be down to problems of survival and transmission, but the inherent challenge to law and order presented by outlying communities of fugitive slaves would have made them a target for royal condemnation and reaction, as well as the ire of ecclesiastical authors concerned with upholding the social hierarchy. It is notable, therefore, that no mention of such communities exists. Placenames in the north and east of England which derive from leysingi, an Old Norse term for ‘freedman’ (for example Lazenby, Lazencroft) could potentially indicate settled groups of fugitives who crossed from English into Danish territory or communities of freedmen who welcomed runaways. However, these settlements are just as likely to be comprised of slaves freed by their owners in order to cultivate more distant land holdings without requiring direct oversight.Footnote ⁴⁴ The precise set of customs necessary to obtain leysingi status make it unlikely that fugitive slaves residing there would have been able to hide their identity for long.Footnote ⁴⁵

If maroon communities were deemed the purview of local lords, this could also explain their absence in texts. However, early medieval runaways had varying options in periods of political fragmentation. Captives generally had better opportunity for escaping their imposed unfree status by returning home, and political enemies or rivals appear to have provided safe havens for fugitives of all types, not just slaves. A lack of recorded maroon communities may also speak to a more general integration of chattel slaves into close-knit, local communities consisting of free and unfree people, which may have abated the need for enslaved people to create their own communities outside the status quo. That said, a lack of evidence for clandestine, peripheral settlements of fugitive slaves does not mean we can discount their presence altogether, as small enclaves of runaways in secluded areas would be difficult, if not impossible, to trace both textually and archaeologically.

Slaves in Wessex may also have found temporary reprieve within churches, where they were granted protection from violence within the sacred space and where ecclesiastics could play a role in mediating punishment with slave-owners. Sanctuary was an ecclesiastical privilege reinforced by secular law which allowed fugitives fleeing criminal punishment and private vengeance to seek asylum within a church’s sphere of protection for an allotted time. This allowed the clergy to enforce a truce period in which negotiations for a peaceful settlement could be held.Footnote ⁴⁶ By guaranteeing the right of sanctuary, rulers may have been able to solidify their reputations as pious protectors.Footnote ⁴⁷ At the same time, ecclesiastical figures were able to extend spiritual protection and provide opportunity for the fugitive to do penance.Footnote ⁴⁸ While it is difficult to perceive this type of escape and negotiation as ‘resistance’, it does illustrate the problems and questions that rulers and ecclesiastics faced in a society which counted some humans as chattel, such as the extent of acceptable mistreatment and whether some slave flight was justified.

On the Continent, the ability of slaves to claim sanctuary in the face of excessive abuse extended as far back as the Theodosian Code and had a strong tradition in Frankish legislation.Footnote ⁴⁹ Early English law was somewhat less explicit. The West Saxon laws of Ine appear to extend sanctuary protection to slaves by stipulating that those who are subject to the lash as punishment for a crime could avoid this by fleeing to a church.Footnote ⁵⁰ Whipping was commonly reserved as a penalty for slaves, presumably because most slaves had no funds with which to make the compensation payments required by people of other statuses.Footnote ⁵¹ Ine §5.1, therefore, differs from Continental tradition by extending sanctuary only to those slaves pursued for a crime, presumably by people other than their owners; it does not offer refuge from abusive masters.

This extension of protective power does not appear to have ever been codified in law, but the tenth-century saint’s life of St Swithun of Winchester, within the former kingdom of Wessex, indicates that sanctuary could be used as a tool to promote benevolent and merciful slave-ownership by allowing for the renegotiation of the relationship between a master and their slave. In Lantfred of Fleury’s Translatio et Miraculi S. Swithuni, composed in the 970s, the author relates several stories involving the miraculous unchaining of slaves awaiting or undergoing excessive punishment by their owners. In each, Swithun posthumously responds to the slaves’ prayers by releasing them from their bonds, allowing the slaves to seek asylum at the saint’s shrine. Upon seeking her out at the shrine, one slave-girl’s furious master was inspired by God to forgo any punishment for her offense or her flight.Footnote ⁵² Other instances of slave flight to the Old Minster do not clarify what became of the slaves after they were released from their bonds at Swithun’s shrine, though one clarification by Wulfstan of Winchester, who wrote his Naratio Metrica de S. Swithuni approximately two decades later, insists that one slave girl’s mistress had tracked her to the minster, and so was present when the miraculous transportation to the shrine was revealed.Footnote ⁵³

At face value, Lantfred and Wulfstan certainly wished to advertise that Swithun’s sphere of protection included the enslaved, the most powerless members of society. But these examples of bad behaviour by slave-owners served as a trope by which hagiographers in early medieval western Europe could reinforce the legitimacy of good, merciful lordship.Footnote ⁵⁴ Though they may reveal incidental details of perceived resistance, Lantfred and Wulfstan sought to leverage the concept of sanctuary to moralize top-down status relations. They may have even sought to promote running to a church as an alternative to outright flight for slaves seeking respite from cruelty.Footnote ⁵⁵ These concerns should by no means be conflated with any sort of abolitionist sentiment, since Swithun’s role as intercessor endeavoured to remedy extreme punishments; neither he nor the living clergy attempted to actively interfere with master-slave relationships, which would have upset the social order and infringed upon a legitimate form of property ownership in which the early medieval Church had significant stakes. Slaves were provided with temporary respite only through ecclesiastical sanctuary and prohibitions on violence within churches. It was only when slaves had committed a crime that the clergy, under the West Saxon laws of Ine, could take an active role in negotiating appropriate punishment and compensation.

Slave-owners perceived flight as a problem throughout the early medieval period. Even the threat of it prompted questions of legal responsibility for the actions of property. It also raised issues concerning the extent to which ecclesiastics could interfere in the master-slave relationship and a slave-owner’s rights regarding the treatment of their own property. While much of what we can identify as slave resistance remains buried within the problems anticipated, even feared, by slave-owners regarding their loss of property, some of this discussion, such as slave flight to enemy territory, seems to have been grounded in reality. These examples illustrate actual attempts by slaves to escape their status which in turn shaped peace negotiations and political relationships. Slave flight as a form of theft may also have played a significant role in the development of the duty to pursue as a legal concept. Evidence for slave flight, therefore, indicates that a combination of elite concerns and actual slave behaviour may have made significant contributions to royal policy.

Theft

Like flight, theft by slaves posed the problem of how free people should treat the removal of property by property. In a world which relied upon communal enforcement, secretive crimes such as theft directly threatened the cohesiveness of a community and remained a preoccupation of law codes. Given that many codes were royally promulgated and theft fell squarely within the king’s jurisdiction, we must be careful to discern whether legal preoccupation with theft by slaves resulted from its severity or merely from the legislators’ need to fully treat theft in all forms.

The latter appears to have taken a guiding role in seventh-century Kentish laws, in which Æthelberht §90 (dated c. 600) and Wihtred §27 (695) both punish slaves for theft, but otherwise appear within lists of related crimes. Æthelberht §90 follows a series of laws specifically targeting crimes against other peoples’ unfree servants (§85–89) and serves as clarification following laws regarding theft committed against the king (§4), by a freeman against a freeman (§9), and against a slave (§89).Footnote ⁵⁶ While Æthelberht §90 importantly portrays slaves as perpetrators rather than victims, this is perhaps only a semantic distinction. Offenses against slaves were considered primarily to be offenses against the slave-owners; conversely, slave-owners were held financially responsible for their slaves’ thefts. We should then view Æthelberht §85–90 as a sequence of laws outlining the rights and responsibilities of lords regarding their unfree dependents which also sought to address a specific theft scenario. A similar theme is present in Wihtred §27, which stipulates the compensation owed should the slave-thief be released, as well as the value owed to the slave’s owner should the slave be executed.Footnote ⁵⁷ This law also comprises part of a sequence regarding the apprehension and punishment of thieves of different statuses (§25–27). Wihtred’s treatment of theft is generally harsher than his predecessor’s, with execution and penal enslavement serving as punishments in addition to compensation payment, but this is just as likely to reflect Wihtred’s and his councillors’ perception of royal authority as an attempt to crack down on rampant thieving by slaves and others.

Tenth-century English law provides a more nuanced view of slave theft, attributable to the changing nature of such codes from records of general oral law into reflections of active law-making, as well as greater focus from the tenth century on the royal oversight of theft.Footnote ⁵⁸ Though part of broader trends, IV Æthelstan §6 (924 × 939) suggests legal generalizations based on expectations of slave rebelliousness. This law details the brutal means of execution due to free people and slaves who commit multiple counts of theft.Footnote ⁵⁹ While this generally only places theft by slaves on a spectrum of statuses, the treatment of female slaves in §6.7 bears special attention, and is worth stating in full:

The law is careful to outline where the treatment of female slave-thieves differs from their male counterparts, who, according to IV Æthelstan §6.5 and 6.6, are to be stoned by eighty of their fellow slaves, who in turn owe three pennies apiece to the slave-owner.Footnote ⁶¹ The additional exemption of theft against the slave-owner is not stipulated for servi. This disparity could simply reflect medieval attitudes regarding the milder treatment of women with fewer circumstances constituting a crime, or it could be an intended inference in §6.5 that theft by any slave against his or her master was not under the jurisdiction of royal law. However, §6.7 states clearly and precisely the ways in which punishment for ancillae should mirror that for servi, making the caveat regarding the victim even more distinctive. As such, it may hold implications for illustrating the gendered roles of slaves and the types of resistance expected of them. Æthelberht’s early seventh-century law code refers to female ‘grinding slaves’ (OE grindende þeowa), and the tenth-century will of Æthelgifu includes the bequest of an enslaved dairymaid (OE dæge).Footnote ⁶² By performing tasks related to domestic production, women were continuously in close proximity to, if not in charge of, desirable but relatively plentiful goods such as food, fuel, and household tools or vessels. Female slaves may have been exempt from royal punishment for theft from their owners not only due to the slaveholder’s right to punish his or her chattel, but because execution would not have been an economically feasible punishment for regular pilfering.

We may catch a glimpse of this gender dynamic in a ninth-century Irish context in the Vita prima Sancti Brigitae, which describes St Brigit’s youth, possibly as a slave, in her father’s household. Brigit’s hagiographer frames her redistribution of goods as acts of charity; what she takes she gives to those less fortunate, and many items are miraculously replaced or replenished. However, not all view these acts as wonderous, and Brigit’s father ultimately attempts to sell her because of what he perceives as her regular thefts from his household.Footnote ⁶³ Here, the circumstances relatable to a ninth-century audience are significant. Brigit steals items which are easily accessible in the course of her duties: meat she is asked to prepare for a guest, butter she has churned, even her father’s sword when it is left in her care.Footnote ⁶⁴ The position of enslaved women as domestic servants in early medieval households may simply have facilitated their theft of household goods. If men, conversely, fulfilled the need for manual labour outside the home, valuable items such as livestock, metal tools, and even building supplies would have been harder to conceal.Footnote ⁶⁵ Æthelstan’s law, then, in seeking to treat all forms of theft as part of changing royal prerogatives, may inadvertently reveal actual or perceived gendered types of slave resistance in tenth-century England.

Nowhere is this push to treat theft and its implications for understanding slave resistance clearer than in III Edmund §4. This code, promulgated 939 × 946, is particularly concerned with property rights as part of the tenth-century drive by kings to establish frið through the legislation of theft.Footnote ⁶⁶ As such, the short code touches upon theft by all statuses, but section 4 relates specifically to slaves who commit theft in groups. Upon capture, the leader is to be executed, and the others are to be scalped and to have their little fingers mutilated.Footnote ⁶⁷ This is the first English law to address theft by groups of slaves and, as the only clause in the entire code without precedent, insinuates certain fears by the elites.Footnote ⁶⁸ While it may only represent an attempt to treat theft in all forms, the specificity and innovativeness of this law suggest real-world inspiration, perhaps reflecting a case known to Edmund. These slaves are not named as runaways, though such a sense may be implied, or the vagueness was created intentionally to suit multiple interpretations. In any case, the law appears to refer to slaves working together to commit and conceal theft, perhaps even within a single estate or across neighbouring landholdings. It may be no coincidence that the two immediately subsequent laws of III Edmund deal with the purchase and pursuit of stolen cattle.Footnote ⁶⁹

Unlike earlier laws, slave-owners in III Edmund §4 hold no liability for the actions of their property, perhaps to account for instances in which a slave-owner might wish to seek some restitution from a group which included their own slaves. By omitting the slave-owner from the list of those responsible, royal law appears to address the complications inherent in punishing the theft of property by property. It also indicates the ways in which slaves threatened or were perceived as threatening communal peace. To ensure that frið was established, kings were forced to consider and introduce conventions for rebellious slave behaviour, which itself indicates that elite fears of slave resistance held an important role in shaping legislation over the course of the early medieval period.

Challenges to Ownership

Ownership of human chattel could also be challenged, both in the context of enslaved peoples’ bodies and in that of their legal person. Early medieval slaves, like those in any time or place, had little to no protection from the sexual advances of their owners.Footnote ⁷⁰ Where early English law criminalized sexual intercourse with someone else’s slave, these restrictions can be seen as efforts to protect the property rights of slave-ownership through control of access.Footnote ⁷¹ These laws also sought to restrict slaves’ liaisons and suggest that the threat of circumvention discomfited slave-owners. Worries about expressions of body autonomy through sexual access or abortion remain folded into broader secular and ecclesiastical reactions to such perceived deviance by women of all statuses. Handbooks of penance, which were particularly interested in sexual sin, address issues such as extramarital sex and abortion even in the earliest examples. The late-seventh- or early-eighth-century Penitential of Theodore, in drawing from the sixth-century Irish Penitential of Finnian, prescribes penance to a woman who terminates a pregnancy, with greater penance required for later stages.Footnote ⁷² While this canon indicates disquiet regarding any woman’s ability to terminate an unwanted pregnancy, this was very likely to include enslaved women, who were sexually accessible to their owners.

Lords and lawmakers may also have been faced with challenges to status brought by captives or their families via legitimate mechanisms. While evidence of this kind is limited, the late-seventh-century Kentish laws of Hlothere and Eadric outlined a legal procedure for cases in which a kidnapped free man returned to accuse his captor, a situation which may have covered those deemed to have been wrongfully sold into slavery.Footnote ⁷³ Regarding status disputes across political borders, we know from one early-eighth-century letter between Brihtwold, archbishop of Canterbury, and Forthere, bishop of Sherborne, that elite families could petition for episcopal help in negotiating the ransom of captive kin, especially when the captive in question was held in servitude by a member of the clergy.Footnote ⁷⁴ Where they were able, then, captives and their families in early kingdoms could harness existing legal procedure and personal networks to negotiate their release.

The general dearth of case law may also leave us blind to the sorts of challenges to status visible in contexts other than wrongful enslavement. Carolingian examples demonstrate how and under what circumstances these challenges could have occurred, with most disputes featuring tenants forced to defend their free status in the face of lords’ attempts to ‘reclaim’ dues and obligations supposedly owed by the tenants’ ancestors.Footnote ⁷⁵ However, one late-eighth-century formula from the monastery of Murbach indicates that the mancipia strove to take advantage of their landowner’s legal conflict. In the confusion of the Alemannic uprising of the 740s, the unfree tenants were able to claim free status with some degree of success, and the monastery lamented the loss of these dependents in a petition to Charlemagne some decades later.Footnote ⁷⁶ This situation further highlights the ways in which unfree people, including slaves, might use conflict and its associated disruption to seize rights and privileges otherwise denied to them.

The issue of people as property prompted significant questions of legal responsibility and social interaction. Even when attention to slave behaviour is directly linked to royal concerns regarding frið, slave resistance was perceived as enough of a threat to the status quo to merit complete treatment in law codes. In the case of VI Æthelstan, elite fear of slave flight as a form of theft may have even driven innovations such as the obligatory nature of a ‘duty to pursue’ thieves. Responses to the problem of people as property not only reveal potential details of slave resistance such as gendered types of theft or a lack of maroon settlements, but they also indicate that elite concerns over slave behaviour became folded into legislation and administrative thinking.

Collective Violence

The complete absence of recorded slave revolts for the early medieval period is hardly unique to England, as the uprisings of the sort led by Spartacus and Nat Turner which had devastating, long-term effects were uncommon, desperate and spontaneous acts known predominantly from large-scale slave societies.Footnote ⁸⁹ The only early medieval example, the Zanj revolt (868–83), occurred near Basra in the Abbasid Caliphate and caused enormous and long-lasting political disruption and resulted in tens of thousands of casualties when a handful of Arabs led East African slaves (the Zanj) to rise up and establish their own administrative system.Footnote ⁹⁰ This rebellion punctuates an otherwise glaring lacuna of documented slave rebellion between the fifth and sixteenth centuries.Footnote ⁹¹

This omission cannot be entirely blamed on the nature of our source material. The social and political commentary which would have been inspired by even a much smaller uprising is precisely the sort which would have interested early medieval authors. A revolt by slaves generally would have been viewed as a shocking act of barbarity, though one by Christian slaves against pagan masters would have been seen as inspired by God. How a ruler handled an uprising would also have served as a barometer for that ruler’s effectiveness. However, even during the transport of captives, a situation which we can presume provided the greatest opportunity for groups of slaves to rise against their captors and escape, no record survives of a rebellion.Footnote ⁹² Nor do we see what could be taken for reactionary calls for limitations on slave movement that might follow an uprising, omissions which are striking and significant.

There is perhaps no sharp distinction between the typical motivations for a slave revolt and an early medieval peasant revolt; both occurred largely in reaction to labour obligations and demands, and free peasants in some cases were spurred by increasing subjection which made them less free – in other words, their personal status. The lack of racially-based slavery in early medieval Europe and the variety of unfree status gradations mean that dichotomous categories of ‘free’ and ‘slave’ which were the focal points of historic slave revolts are not as sharply delineated as in pre-modern and early modern large-scale slave societies, our main points of comparison for slave-led rebellions. Many early medieval popular revolts, however, were not motivated by interests which could be shared with chattel slaves or the hereditarily unfree. In the Stellinga revolt, for example, the rebels were foremost driven by newly imposed labour obligations, and the tax revolts of eighth- and ninth-century Egypt were inherently related to free status given that slaves did not pay taxes.Footnote ⁹³ If peasants in early medieval Europe were able to mobilize for collective violence in response to perceived oppression, it is worth querying why the enslaved did not do this.

This line of inquiry takes on a special note of importance for England, which had neither slave nor popular revolts. Approaching slave rebellion, an expectation of elites in many slaveholding societies, from a distinct angle can provide us with a better understanding of how slavery and unfreedom functioned in early medieval England and how these institutions compare to other historic contexts. The set of conditions most adaptable to a medieval situation was identified by Eugene Genovese in his 1979 monograph, From Rebellion to Revolution, and was developed initially to query the greater frequency of slave revolts in Latin America and the Caribbean than in the American South. This influential methodology has subsequently been refined by David Geggus from a global perspective, and by classicists who have specifically adapted Genovese’s indicators to the ancient world.Footnote ⁹⁴ In summation, Genovese recognized the following criteria as significant:

1. 1) a master-slave relationship characterized by absenteeism, with slave-owners in Europe rarely or never visiting their colonial plantations, as well as depersonalisation and estrangement between resident free people and slaves;

2. 2) economic distress which prompted famine in the localities and engendered desperation amongst slaves;

3. 3) large slaveholding units in which approximately 100–200 slaves laboured in close proximity;

4. 4) a ruling class consistently divided by political struggle within or with another slaveholding polity, and thus distracted from quelling unrest amongst slaves;

5. 5) slaves and freed slaves greatly outnumbering those born free, in order to combat the military advantages of the free population;

6. 6) enslaved captives outnumbering those born into slavery, as captives were more likely to resist their situation;

7. 7) the accepted presence of autonomous leaders amongst slaves, who could become leaders of a movement;

8. 8) and conditions for the successful establishment of maroon settlements, which could provide external assistance to rebelling slaves and militarily threaten the slave system.Footnote ⁹⁵

Genovese stressed that these criteria were neither all-encompassing nor infallible, and he presented them in no order of importance. Classicists such as Paul Cartledge and Theresa Urbainczyk have only made minor adjustments to the model, exchanging racial terminology for status indicators (as I have done here, for example, replacing ‘Black’ with ‘slave’ as appropriate). This application has resulted in detailed and plausible hypotheses for why ancient Greek and Roman slave revolts occurred when and where they did. Geggus, on the other hand, has interpreted large slaveholding units and a high ratio of slave to free as being the only direct causes of rebellion, believing Genovese’s other criteria, along with potential factors such as increased freedoms amongst urban slaves and the presence of emancipationist rumours, to be inconclusive. While urban slaves may have had wider networks and greater freedoms, increased military presences in urban areas may have pre-empted revolts, and assertions by rebel leaders that they were fighting for freedom they believed was being withheld by a local authority does not illuminate the degree to which slaves actually believed these claims.Footnote ⁹⁶ Geggus only challenges the importance of maroon communities, arguing that marronage existed as an alternative to rebellion, not a catalyst for it.Footnote ⁹⁷

Even with modifications and critiques, however, Genovese’s criteria serve as a significant benchmark by which we can establish that favourable conditions for a successful slave uprising did not exist in early medieval England and that the lack of concern by medieval authors accurately reflects an absence of revolts. Considerations of slave population size and group ratios, deemed fundamental by Geggus, can be immediately discarded. Slaves in early medieval England comprised a relatively small, dispersed population which was likely overseen directly by slave-owners or free members of society. Where the problematic statistics gleaned from Domesday Book estimate England’s slave population to have been approximately ten to twelve per cent in 1066, in Latin American and Caribbean colonies, where violent revolt is well-documented, slaves could constitute upwards of eighty per cent of the population.Footnote ⁹⁸ Even if we consider that slavery in England may have been a dying practice by 1066, there is no evidence to suggest that slaves comprised as substantial a proportion of the population there as they did in the Americas. Medieval slaves assigned to agricultural labour, therefore, were probably not concentrated in large groups separate from the free labourers where they could meet, plan, or gather support. It is notable that despite the enormous numbers of people enslaved throughout the medieval Islamicate world, the Zanj revolt occurred in what may have been the only part which utilized large-scale agricultural slavery.Footnote ⁹⁹ As a relatively small population which may have been intermingled with free and other unfree tenants on estates or even resident in their owners’ homes, slaves in early medieval England had greater potential to form personal relationships with people of other statuses, which may have forestalled violent uprising against them; certainly the free and the unfree were not estranged groups. Such a situation also enabled closer, more consistent oversight which was not conducive to planning group acts of resistance, and the lack of any substantial maroon settlements would have negated the potential for external assistance in raising a revolt. The lack of uprisings in transit would likewise insinuate that even during periods of more expansive slave trading, captives and slaves were not transported in extensive caravans or crowded onto ships in numbers exceeding that of the crew.

The nature of slavery in early medieval England, then, fails to meet half of Genovese’s criteria and none of Geggus’s direct factors in facilitating slave rebellion. The paucity of information makes it difficult to establish whether societies allowed for the emergence of slave leadership, or whether enslaved captives outnumbered those born into slavery.Footnote ¹⁰⁰ We can at least be confident that throughout the early medieval period economic distress and famine occurred, and that the upper echelons of the slaveholding hierarchy were frequently divided by conflict either amongst themselves or with neighbouring slaveholding societies. By the aforementioned criteria, an image arises of a slaveholding society which lacked many of the principal attributes necessary for fomenting and sustaining a substantial slave revolt.

If unfree people did express collective, violent dissent, the groups and the consequences must have been considerably limited in size and scope: numbering only a handful of people and restricted to a small geographical area, perhaps a single village or estate. Records of such unrest, particularly on secular-held lands, could easily have been lost or omitted from surviving texts. They would have either fallen under the slave-owner’s jurisdiction for punishment, provided no outsiders were injured and no damage was done to another landholder’s property, or within the general prescriptions of extant legal procedure for injury and property damage. Even so, such uprisings would have to have remained extremely small to escape the interest of royal authority or ecclesiastical moralisers.

Despite differences in the detectable motivations of slave and early medieval popular revolts, the absence of both in England could be explained by the same social conditions and legal norms. With a few rare exceptions, communities sought to redress their personal and collective grievances regarding dues and services with individual lords as the immediate controllers of these obligations rather than relatively weak early medieval states, and thus large-scale revolt was not considered a serious and consistent threat to elites until it happened in the fourteenth and fifteenth centuries.Footnote ¹⁰¹ The reasons why nothing like the English Peasants’ Revolt of 1381 occurred before or after that date remain difficult to pin down, though it may have resulted from conditions unknown in the early medieval British Isles, such as rising peasant expectations and resultant tension with lords in the aftermath of the Black Death, or in a culmination of changes in English administration by the late fourteenth century, such as greater centralized control of the periphery through the facilitation of communications, increased jurisdiction, and developing systems of extraction.Footnote ¹⁰² Perhaps only then could peasant discontent achieve a broader geographical scope and move beyond individual disputes with lords.

Individual Acts of Violence

Likewise striking in their invisibility are individual acts of violent resistance, though in most cases the limited scope of such an act means they were less likely to be recorded. One Irish example makes clear that specific events, such as the murder of a slave-owner, were only worth documenting when they held great political significance; it is for this reason that the Irish annals recorded that in 989, Glúniarann, the Viking king of Dublin, was killed by his own slave while drunk.Footnote ¹⁰³ Normative laws do identify the act of homicide by unfree servants as enough of a threat to justify legislation, though only through the early eighth century.Footnote ¹⁰⁴ The early Kentish laws of Æthelberht dictate the compensation due by one esne who kills another who has not committed an offense, and those of Hlothere and Eadric extend this to the examples of an esne killing a noble (OE eorlcund) or a freeman whose wergild is 100 shillings.Footnote ¹⁰⁵ In roughly-contemporary West Saxon tradition, Ine §74 lays out punishment should a Welsh slave (OE ðeowwealh) kill an Englishman, adding to the general concern over homicide by slaves the apartheid-like, anti-Welsh dimension for which his law code is well-known.Footnote ¹⁰⁶

Regarding violence more generally, the weapons restrictions of Ine §29 have already been discussed in the context of flight. This law was as much intended to protect the safety of slave-owners as it was their property investment. The aforementioned Irish example in Bretha Éitgid must have also reflected slave-owners’ concerns that slaves could intentionally injure people or livestock.Footnote ¹⁰⁷ While these laws do not necessarily reflect violence against slave-owners themselves, they do indicate a perception by slaveholding elites that free people needed protection from and compensation for violence by the enslaved.

Suicide, being violence against oneself, was a significant feature of other slaving systems, a drastic action taken by captives and slaves to avoid what they perceived as a fate worse than death. During the Middle Passage, enslaved Africans frequently leapt from the ships in response to the traumas they experienced and anticipated, to the extent that slavers shared advice for suicide prevention.Footnote ¹⁰⁸ Even if depictions by Roman authors and sculptors of slave and captive suicide constitute a romanticized trope, slave suicide was certainly a point of practical discussion in Roman law.Footnote ¹⁰⁹ No mention of this practice exists for early medieval England or its neighbours, but as discussed above, captives generally had better chances of returning home or fleeing to another political territory. Furthermore, Christian society in the Middle Ages perceived suicide as a deviant, sinful measure. This type of death had been sharply denounced by St Augustine of Hippo in his early-fifth-century De civitate Dei, and a similar ecclesiastical attitude very likely extended into the early Middle Ages.Footnote ¹¹⁰ Early medieval burials in England at boundaries and roadsides may in part represent suicides who were denied interment on consecrated ground.Footnote ¹¹¹ We can expect that any cases of captive suicide, even by early pagans or non-Christian vikings, would have been actively omitted from the records made by Christian, usually ecclesiastical, authors. The implications of suicide for one’s eternal soul imply that it may have been rare amongst Christian slaves and captives. Even if slave-owners did have concerns regarding suicide, cultural taboo appears to have rendered these worries invisible.

The evidence of individual acts demonstrates that slave violence remained a serious concern for slave-owners, and that such behaviour could still force questions of responsibility and criminality. Some acts for which we have evidence can very clearly be classed as forms of slave resistance – the killing of one’s owner, injury to those people preventing escape, or injury to the livestock crucial to their labour. Others, such as the killing or injury of people generally, have no intimation of intent and tell us more about inter-status relations than about slaves’ perception of their own status. Though they are evidence of a negative, the lack of slave revolt and slave suicide, common aspects of large-scale slave societies, serve to underline the nature of early English slaveholding both in its relatively limited scope and the social conditions within which it operated. Communal justice and integrated communities of free and unfree people may have pre-empted slave revolt in the same way that they seem to have forestalled popular uprisings, though why the free peasantry responded to increasing aristocratic power seemingly without collective violence has yet to be understood.