This essay has two purposes. The first is to examine the ways in which slavery was defined in the medieval and early modern ius commune. The second is to examine some of the consequences of that definition for European men and women. The sources of the essay are largely confined to the academic law, the texts and the treatises of the medieval schools. However, legal theories almost always have real world consequences, and of the inevitable connection between law and fact, the history of slavery’s definition provides many illustrations.
The ius commune was the amalgam of Roman and canon laws that dominated European legal education and shaped practice in European courts between the twelfth century and the nineteenth. It deserves the attention of anyone who seeks to understand slavery’s history. It was the vehicle through which the classical Roman law reached the modern world, and it furnished the principal means of defining slavery during the Middle Ages. It did leave room for local variation through the force of custom and statute, but the ius commune exercised control over legal thought and much of legal practice throughout Western Europe from the revival of legal study at Bologna during the twelfth century to the age of Codification.1 It shaped what lawyers thought about slavery. If, as we shall see, it was not the final word, for centuries it provided the starting point for the treatment of slavery in European courts. Examination of the ius commune therefore provides one instructive way of understanding the larger history of the subject.
Many of the definitional problems that confronted delegates to the Slavery Convention of 1926 and its successors also confronted jurists within the traditions of the ius commune. The 1956 addition to the list of ‘practices similar to slavery’ in the Supplementary Convention’s coverage in fact raised a very old problem, one the medieval and early modern commentators themselves had to face.2 They grappled with it. Of course the surrounding circumstances were quite different for them. So were many of the conclusions they drew about slavery itself. Under the ancien régime, slavery was not treated as contrary to the law of nations, as it is today. The jurists therefore began their work within a climate of opinion that was moving only slowly away from acceptance of slavery as an inevitable part of human life. However, they faced the need for definition nonetheless. How to define slavery? What restrictions on human freedom counted as true slavery, leaving the slave without legal rights? What legal disabilities and what legal conclusions followed from the definition adopted? These were live issues in the ius commune. The conclusions reached by the commentators of earlier centuries, though not our own, remain illustrative of the attitudes towards slavery of thoughtful men. They provide an essential chapter in European slavery’s long history.
There is one additional reason for taking up this subject within the general history of slavery. Many historical works about slavery skip over the medieval period.3 There are exceptions, of course, the outstanding examples being Charles Verlinden’s classic work and the more recent studies of Emanuele Conte.4 However, the former’s geographical approach left little room for exploration of the ius commune, and many general histories of slavery that concern themselves with the law on the subject have had little to say about it or about the Middle Ages. Their authors sometimes note its theoretical survival in law, but they do so only briefly. Beginning with the Roman law of slavery, they make only a quick stop in 1500 before moving to the New World where the real contest over slavery’s legitimacy would take place.5
Justifications for this neglect can be given, of course. By 1500 slavery as an economically important institution had disappeared from many parts of Europe, as jurists of the sixteenth and seventeenth centuries themselves recognized.6 Even so, personal servitude was not dead, only diminished in scope and changed in form.7 Hugo Grotius (d. 1645), for example, admitted that ‘traces’ of slavery remained in the law of Holland during his own day.8 English documents from the early sixteenth century also included forms for manumission of unfree men and women, apparently meant for use.9 Treatises devoted expressly to the law of servitude were written in the sixteenth and seventeenth centuries, something that would not have happened so often had the subject been wholly obsolete.10 Some of them dealt with the question of whether the many servants who peopled early modern society were to be classed with ancient slaves for legal purposes.11 Similarities existed, although the differences usually seemed as substantial as they did. From such works a tradition that led to the modern law of master and servant gradually emerged. To neglect slavery’s status in law for so long a period of time leaves out a great deal. The ancient institution still existed in many different forms. It was not only because slavery was enshrined in their texts that the jurists were obliged to deal with it.
Except in England, the education of lawyers began with the texts of the Roman law, and that is where we should begin too.12 The Corpus iuris civilis treated slavery as a living institution, and it opened with what seems to be a firm, even unbridgeable, dividing line between men who were slaves and those who were free. ‘All men are either free or slave’ (Inst 1.3.1). It must be the one or the other. They could become subject to slavery in different ways, but ‘the condition of all slaves is one and the same’ (Inst 1.3.4). So said the foundational texts.
Reading and seeing these words and finding them oft repeated in texts throughout the Digest must have made a stark and uncompromising introduction to the law of persons. It announced the strongest sort of division between the slave and the free. It proclaimed the reality of slavery’s existence. Working through the Digest, however, students would have discovered that the matter could be more complicated. Slaves could become free by manumission. Slavery was not always an indelible mark. If freed, they occupied a kind of ‘halfway’ position; under certain circumstances, they could be returned to slavery. Freedmen remained bound by several duties towards their patrons. Even in cases of emancipation, therefore, slavery was very much a fact of Roman life and law, not an aberration.13 Students would have confronted that fact from the start.
The student of the law of the medieval church, the canon law, would not have found so explicit a treatment of the law of persons in the Corpus iuris canonici, but neither would he have seen the civil law contradicted by the canonical texts. The aspiring canonist would have learned much the same lesson as did his civilian counterpart. Slavery existed. It had been introduced into the society of men by the law of nations. Drunkenness, personal indebtedness, and war may have been the immediate causes for its institution,14 but their exceptional character did not render slavery unlawful. Slavery was not itself contrary to the laws of the medieval church.15 The Bible mentioned slavery without condemnation, and any alert student of the church’s law would also have known of its contemporary existence by experience. Medieval churches themselves possessed slaves. For the canonist as for the civilian, slavery would have stood as a fact of life.
In assessing the impact of this legal recognition, two features of medieval life should be remembered. They stood in tension the one to the other. The first is that neither the law of the Romans nor their medieval successors treated freedom from personal restraint as an unambiguous good. The jurists did develop an understanding of civil rights, according a place for them in their legal systems,16 but except in special circumstances it was not a reflexive reaction of medieval thinkers to assume that all men and women should be entirely free to do as they pleased. The Roman law’s patria potestas accorded a very large measure of control to fathers over children, even adult children. However softened that stern regime became in the course of the centuries, it remained the starting point for the jurists of the ius commune. They did not think of all abridgments of freedom as slavery, and they were not slavery in fact. However, the slavery of which they read was only a harsher form of control over persons than that which permeated the law and medieval society generally. Indeed the legal status of the person serving what contemporaries regarded as the highest calling—monks and nuns—lived under a regime not too far removed from actual servitude. To be a ‘slave to God’ was altogether praiseworthy.17 The ‘free man’, the person bound by no obligation to others, was likely also to be the ‘friendless man’. He was not to be envied. His condition was not a model to which men should aspire.
The second is that, although permitted under the ius gentium, slavery was contrary to the law of nature. True, there was doubt about the origins of slavery, but the relevant legal texts stated the natural law’s precept clearly, and where they had a choice, jurists were directed by natural law to follow a presumption in favour of freedom. With this presumption, they rejected the idea found most clearly stated in Aristotle, that slavery was part of the natural order of things.18 Sometimes, the jurists put this in striking (though in fact conventional) language. ‘Slavery is equivalent to death’, wrote Cardinal Tuschus (d. 1610), ‘and just as it is lawful to take any means to evade death, so is it lawful to avoid slavery’.19 Strong statements like this one, which was actually lifted (out of context) from the Digest,20 quickly met reality. Although contrary to natural law, slavery was not unlawful in the sense that a court could decide that it did not exist under the law; no judge could ‘strike down’ slavery as a modern American judge can ‘strike down’ an unconstitutional statute. However, in practice, the natural law’s stance on slavery did make a difference in the decision of legal disputes. For instance, wherever possible statutes were to be interpreted to restrict the scope of slavery. If there was doubt about the meaning of a statutory term, it was to be interpreted to promote human freedom.21 This was the favor libertatis of classical law.22 It could, and sometimes did, play a role in the application of the law of slavery.
Article 1(1) of the 1926 Convention defined slavery as ‘the status or condition of a person over whom any or all of the powers attaching to the right of ownership are exercised’. The substance of this definition has been repeated and refined more than once, but its completeness and utility have long been matters open to discussion.23 Definition was obviously necessary as part of the effort to end the institution of slavery; one had to be able to determine whether or not it existed in any specific case. In fact, medieval students of the Roman and canon laws would quickly have faced the problem of definition in their own studies. Perhaps it has been open to doubt in all societies that have recognized slavery’s existence.24 If a slightly ambiguous acceptance of the institution marked the start of a student’s understanding of slavery as contained in the law of persons, it was very far from the end of uncertainty. He would soon have discovered that a considerable range of personal servitude existed. As just noted, freedmen remained subject to personal disabilities despite their manumission. They could, for example, be returned to slavery if they proved ‘ungrateful’ towards their former master.25 They were also bound to render an account of their property to their patrons.26 Were they, therefore, to be included in the category of servi? Or quasi servi? If, as it seems, they were free in a formal sense, it was a precarious sort of freedom.
Apart from such limitations on freedom, during the Middle Ages there existed a large number of men who were neither completely free nor completely slave. Classical Roman law had regarded it a logical absurdity to suppose that a man could be part free and part slave,27 but exactly that situation recurred in fact. The law of later centuries had to come to grips with it. For example, there were men known as dediticii: citizens of a vanquished foreign state who had surrendered to the might of Rome. They were technically free but they laboured under the loss of most civil rights. According to the medieval glossa ordinaria, they ‘lived as slaves but died as free men’.28 In addition, the term mancipia was often used as a synonym for slaves in the texts, although sometimes it was also used for free persons.29 Then there was the large class of coloni, originarii, or adscripticii, men and women who were technically free but who were ‘bound to the soil’ and its cultivation.30 They could be ‘reclaimed’ by the owner of the land on which they toiled if they deserted their duties and sought freedom elsewhere. Herein, it is thought, lay the origins of medieval serfdom. Men and women classed as nativi or villani or rustici seem also to have been similarly obligated to service and to the soil. The famulus seemingly held a roughly similar status, except that his service usually occurred within the master’s household; famuli were not ordinarily tied to the land.31 The operarius appears to have been similar, except perhaps in the place of obligatory service. European vernaculars knew a similar multiplication of terms denoting semi-free status.32 Exactly what differences in status were meant by these different terms is not always clear in the sources. Some terms applied to men of servile condition—the ragacinus for example—have proved difficult to trace.33 Perhaps it was only a local name, equivalent to some of the others in substance. The profusion of such names may sometimes actually have confused the commentators themselves.34 It was hard to know exactly where they all fit within the law’s apparently firm classification scheme.
With these categories, seemingly confusing in themselves, the jurists did nonetheless begin their analysis with the lawyerly assumption that different terms must mean different things, and they did accept the larger point that gradations of freedom could exist in law as well as in fact. In one sense they asked the same questions about men and women within these classes as appeared in the 1926 definition: were they subject to any or all of the powers attaching to the rights of ownership? Because slavery in most forms was not illegal in earlier centuries, the consequences of falling on one side of that definition or the other were not the same as they would be later on. However, the answer given to this question did matter in fact, because it often served to determine the legal and social rights of the men and women involved.
The principal subject of this contribution is the illustration and examination both of the question of slavery’sdefinition and its consequences in the lives of those affected by it. This can only be done by exploring detailed legal questions. I have chosen five separate but fairly representative examples of the way the medieval law grew from its starting point in Roman law. Each of the five begins with the categories found in the classical law and moves into the use made of them by the medieval commentators. Some of what one finds was old, consistent with the Roman categories, but some was quite new.
Crimes committed by men of servile condition posed intellectual and practical problems for the jurists who began with the Roman law. No general criminal law existed in the Corpus iuris civilis, and from the mass of specific criminal statutes that existed within it, few clear principles emerged. It was not easy to extract general rules applicable to slaves from that mass. At one time, punishment of a criminous slave may have been left to the slave’s master, but that regime had been brought to an end in the early Empire.35 Even the master’s right to chastisement had been limited by imperial decrees.36 Some difficult questions therefore arose when slaves committed crimes, and that happened often enough to require a more developed law than that found in the Digest.
The arguments for each side were reasonably well balanced in the texts. On the one hand, to punish a slave who had committed a crime could amount to punishing the slave’s master. The master would suffer from loss of the slave’s labour. For the former, that would amount to punishment without fault, and perhaps there was something in the argument against punishing the slave at all if the punishment would be felt principally by the innocent master. It did have some traction in the canon law.37 On the other hand, the principles of interpretation applied to criminal statutes by the classical and medieval jurists meant that the operative texts did seem to apply to slaves, at least in the absence of a contrary intent expressed in their language. The jurists therefore rejected the notion that the master should always stand between the slave and punishment under the criminal law. The reason seemed self-evident. Not to punish a slave who had committed a crime would license unlawful behaviour. Indeed some crimes—the servus fugitivus who ran away from his master for example—could be committed only by slaves.38 Under these premises, it was clear that slaves could be punished in some circumstances consistently with the texts in the Digest.
Problems recurred, however. Punishment for some crimes would evidently have taken the master’s property rights, punishing him more than it would have hurt the slave. Depriving the master of work done by servi might actually make the latter better off. He would not have to work. The none too elegant ‘solution’ to the problem was to allow a criminal accusatio to be brought against a slave except where the penalty attached to the crime consisted of either loss of property or loss of status. In the first case, no penalty could be exacted because by definition the slave held no property. What he had actually belonged to the master.39 In the second case, no penalty could be exacted because in the eyes of the law the slave had no status.40 He was the property of another and therefore a man without legal rights. He himself had no possessions from which a monetary penalty could be taken. It was objected that the slave might have a peculium, the sum of money or property conceded to the slave by the master for the former’s own use. Could not the criminal penalty be extracted from that source? No, it was held, because the peculium should not be taken indirectly from the person to whom it truly belonged, the master.41