Briefly exploring the historiography

1In the same way that European Medievalist and Modernist historians have traditionally disagreed regarding the labour value of slavery, they do not agree regarding the definition of slavery and captivity. In the past few decades American and Anglo-Saxon modern historiographies posit that captivity was a condition equivalent to slavery. Moreover, historians maintain that the captivity of Christians in North Africa is exemplary of Mediterranean slavery. Titles like Christian Slaves, Muslim Masters: White Slavery in the Mediterranean¹ leave no doubt in this regard. More recently, numerous French and Italian scholars have embraced this perspective, as they demonstrate in several studies in important anthologies, such as Schiavitù Mediterranee. Corsari, rinnegati e santi di età moderna,² L’esclavage en Méditerranée à l’époque modern,³ Les esclavages en Méditerranée. Espaces et dynamiques économiques,⁴ or Serfdom and Slavery in the European Economy 11^th-18^th Centuries.⁵ Others have followed a different path, as can be seen in Le commerce des captifs: les intermédiaires dans l’échange et le rachat des prisonniers en Méditerranée, XV^e -XVIII^e siècle⁶ just to name a few of the collections in what has been a significant expansion of the historiography in recent years.⁷

2On the other hand, surprising as it might seem, it has also been argued that slavery was a manifestation of captivity, as Jean Dunbabin asserts in her monograph on captivity in medieval Europe. It is true, though, that the author limited her study to Christian Europe, focusing on prisoners and hostages among Christians, without taking into account the ideological factor that is key in the contacts between Islam and Christendom.⁸

3Convinced of the need to avoid semantic, legal and historical confusion of these concepts, the present study aims to underline that slavery and captivity were, indeed, differentiated conditions, despite the obvious similarities, beginning with the deprivation of freedom and following with the labour profit obtained. I shall insist on this reality on the basis of documents and studies from the fifteenth-century Iberian Mediterranean context (looking specifically at material concerned with the Crown of Aragon, the Crown of Castile, and the Kingdom of Granada), with an eye toward Islamic law in the western Mediterranean.

4The chronological and geographical frame chosen is not unintentional: the differences between slavery and captivity sharpened during the fifteenth century, when the term “slavery” displaced “feudal serfdom” to refer specifically to enslaved human beings belonging to religious communities different from those who deprived them of freedom. But it has been a long time since the question was first tackled: when Charles Verlinden, pioneer in the study of medieval slavery in Europe, dealt with the origin of the term slave, he did not consider that slavery and captivity were different conditions; rather, he pointed out that captive was one among the terms used in the Iberian Peninsula to refer to slaves, thus reinforcing their interchangeability.⁹

5Giulio Cipollone has noted that scholars who work on European medieval slavery did not consider the difference between the terms because captivity was assumed as the logical, previous premise of slavery.¹⁰ His assertion is true, for it has been modernists who have assumed that slavery and captivity were two terms for one phenomenon. The climax of this position is resumed by a widely cited assertion from Michel Fontenay, who recently wrote: “I would gladly say that the captive is a slave waiting to be rescued, while the slave is a captive who does not hope anymore for ransom”, a vision which, taken further, has led the French scholar to assume that “there was, in my opinion, a specific slavery in the Mediterranean in modern times, a slavery among white people, not reducible to other contemporary forms of enslavement”. In the end, from an economic point of view, the main difference between captives and slaves was what he calls “exchange value” and “use value”.¹¹

6From this starting point, defined for the modern Mediterranean, Fabienne Guillen and Salah Trablesi transferred it to the medieval Mediterranean, proposing in the Introduction to their collected volume the need for “a decentralisation of notions, so that, far from assuming beforehand the split between captivity and slavery, we may tackle from a comparative perspective the capture and trade dynamics, and the tax, commercial and fiscal logics, questioning the plurality or singularity of the dominant model of slavery”, in order to talk about “slavery spaces, temporalities, economies and politics”.¹² In the end, as we can see, this reasoning has made both conditions equal, as if they were identical and interchangeable.

7No matter the method, to achieve a critical, substantiated position we should proceed questioning if captivity and slavery responded to the same reality, rather than searching for the reasons why there are two terms to describe a reality which some see as unique. Therefore, in this study I will resume my own reflections from more than a decade ago and further develop them, according to new and more recent studies.¹³ I will broaden the frame to the Iberian Mediterranean context, with the Kingdom of Granada at its centre as the frontier between western Islam and Christendom, always bearing in mind that modern captivity and slavery in the Mediterranean were natural heirs of their late medieval equivalents.

Slavery and captivity, between law and ideology

8Examining medieval law is key to understand the differences between slavery and captivity. The Siete Partidas or Seven-Part Code was a unique code of civil law compiled under the reign of Alfonso X the Learned, King of Castile in 1252–1284, with the intent of standardising the normative rules of the kingdom. Moreover, it is interesting to underline that it includes references to Graeco-Roman, Judaeo-Christian and Islamic tradition.¹⁴ Therefore, it is interesting to confirm that there were different paths that led to slavery. Indeed, the Partidas differentiated three: to be born from a slave mother; the laws of war in the case of infidels; and a free man who accepted being sold into slavery.¹⁵ On the contrary, a man would become (“devenir”) a captive, acquiring the condition.¹⁶

9In the same way, the mechanisms to leave both conditions were also different: captives were freed through redemption, while slaves recovered freedom through emancipation or manumission.¹⁷ Redemption, emancipation and manumission were not, without discussion, equivalent legal terms. To understand the difference among them we must resort to the ideological speech of the times, as Andrés Díaz Borrás did when analysing captivity in late medieval Valencia. This author went back to Jewish and Roman traditions, which later shaped the Christian conception of captivity. Indeed, once they posited the legal difference between a slave and a free man, Romans understood that captives were prisoners of war deprived of freedom by an enemy who had enslaved them. Yet, the ideological element was assimilated from Jewish tradition when Christianity became the official religion of the Roman Empire. As a matter of fact, Judaism differentiated between slavery and captivity referring to a clear, undisputable turning point: the alliance of the people of Israel with Yahweh. Thus, Hebrews were slaves in Egypt yet captives in Babylon: after they sealed their alliance with the divinity, their deprivation of freedom constituted an affront against a religious entity as well as a social and political order.¹⁸

10After Christians inherited this ideological meaning of captivity, the Church did not condemn slavery. And although it admitted slaves in the community of believers, it did make a difference between slaves and captives by keeping the second term to refer only to Christian believers who had been deprived of liberty by infidels. The loss of freedom at the hands of a member of another religious community entailed the danger of apostasy, with the consequent risk of loss of salvation of the Christian’s soul. Yet, this danger did not materialise until the expansion of Islam in the seventh and eighth centuries across the Near East and North Africa made it a real menace. As a result, it is no surprise that Díaz Borrás pointed out that the heart of the problem lies in the antagonism of the two great Mediterranean religions of the Middle Ages.

11From the moment the Church considered that Christians should never be slaves of Muslims due to the spiritual risks of the situation, Islam as an ideological rival caused an evolution of both concepts, slavery and captivity. Therefore, I can only agree with Díaz Borrás that the first is a social concept, while the second is an ideological one.¹⁹

12On the contrary, we have to insist on the difference between terms from a legislative point of view, given that some authors have even stated that the juridical condition of slaves and captives was the same.²⁰ If we resort to Roman law once more, a Roman citizen deprived of liberty, a captive as we have seen, did not have the same juridical condition as a servus (assimilated in its condition to a medieval slave). Likewise, Christian captives in the Middle Ages lost freedom de facto, while slaves lost it de iure.

Muslim captives and slaves

13The ideological and legislative discourses seem more than enough in their approach and internal coherence to make a clear, simple difference between slavery and captivity. Yet, late medieval documents reflect the presence not only of Moorish slaves, but also of Moorish captives, which likewise has to be explained.

14If slavery cannot be discussed as the loss of freedom de iure, it is also true that it is impossible to keep the term captive exclusively for Christians in the hands of infidels. Once more, the Partidas shed some light on the problem with a careful reading of its definition:²¹

Captives are rightfully called those who are imprisoned by men of other faiths, who kill them after they have imprisoned them despising their law, or torment them with cruel punishments or use them as serfs in such a way that they rather wish death than life.

15Therefore, captives were men who had lost freedom at the hands of others of different faith, who might use them as serfs. As for this last term, we have to remember that the Partidas were written in the thirteenth century, when the term slave did not exist yet, although we have to assimilate this serfdom to slavery, which is never mentioned as such through the whole code. In this sense, as Charles Verlinden stated, the word slave did not become part of the Iberian languages until the fourteenth century in quite a restricted way, considering that its use did not actually spread until the fifteenth century.

16Once accepted that a captive was a prisoner in the hands of a man of another faith, there is no reason at all that prevents the use of the same term regarding Muslims deprived of liberty in the Iberian Peninsula. Moreover, if we resume the ideological argument that confers the captive the possibility of changing religion, we reinforce the understanding of the concept, for if a Christian in the hands of a Muslim was a captive because of the risk of apostasy, inversely a Muslim as a captive could also apostatise Islam and become a Christian.

17Yet, this reasoning, which can also apply to the Jews, is not valid for Sub-Saharan black-Africans. Considered as pagans –in contrast to “black Moors”, who were Muslims– and not belonging to any organised religion that the Church would recognise as such – as was the case with Islam – sub-Saharan people were always considered exclusively as slaves, never as captives: there was no religious-ideological factor in their case.

Captivity before slavery: captives that did not became slaves

18If we focus our attention on the Muslim population deprived of liberty in Mediterranean Christian Iberia – that is, the Crown of Aragon mainly, and the Crown of Castile after the final conquest of Granada in 1492 – new questions arise about their status: what was the difference between the Muslims of Malaga, held in captivity after the fall of the city in 1487, and the Muslims enslaved after the conquests of Tripoli, Oran or Tunis? Were the so-called hostages of Daidin equal to the Muslim captives “of good war” (“de buena guerra”)?

19In the Crown of Aragon, mainly the Kingdoms of Valencia and Majorca, as many authors have shown, the difference was very clear: when an enemy of different religion, usually a Muslim, was made prisoner and therefore become a captive, they had to be introduced to the Bayle General, a high Crown officer in charge of the Royal Treasury who, after finding out about the origins of the captive and the circumstances in which he lost freedom, had to determine whether he was “of good war”. In other words, the royal officer, after examining the captive, had to decide if the loss of freedom was legal in the light of war laws that presided over Christian-Muslim relations in general, and if so, declare the captive a slave. Once the captive had become a slave, his market value was estimated to deduct the quinto real, a tax that meant the fifth part of the total value, which the owner had to pay before having the slave totally at his complete disposal.²² Therefore, it seems clear that this procedure confirms that captivity was a state that preceded slavery, in which the person deprived of liberty was held prisoner until his fate was decided: released free in the case of illegal captivity, or declared a legal slave.

20By contrast, in the Kingdom of Castile there was no institution equivalent to the Baylia, a circumstance that made it more difficult to establish a qualitative difference between both conditions.

21To explain this situation Enrique del Pino suggested that Christians had more difficulties when redeeming their coreligionists. A more reduced trade of captives in the Crown of Castile before the incorporation of the Kingdom of Granada would have rendered more problematic the understanding of the nuances that differentiated the captive from the slave during the Late Middle Ages, for their limits would have been rather confusing. According to this author, the changing context at the end of the fifteenth century, with the War of Granada and hostile relationships with North Africa, would have contributed to definitely establish the contrast between both. Yet, he does not specify which were these nuances.²³

22Mateo Páez, who has studied slavery in Cordoba, proposes distinguishing between captives and slaves according to the ills that could be inflicted, as expressed in the Partidas.²⁴ Indeed, as we have seen, the legal code considered that “captives are rightfully called those who are imprisoned by men of other faiths, who kill them after they have imprisoned them despising their law, or torment them with cruel punishments or use them as serfs”, while the slave, in contrast with the captive’s total lack of rights, had some granted, at least in theory, regarding physical integrity, marriage and a limited juridical capacity, which have been pointed out by J. A. Doerig.²⁵ But neither explains how one person might go from one condition to the other.

23Yet, the total lack of rights pointed out by Mateo Páez is illuminating because it allows us to understand the meaning of the expression “His Highnesses’ captives”, that the documents use to refer to the population of Malaga, deprived of liberty after the Christian occupation of the city in 1487. The refusal to surrender to the Catholic Kings in the terms requested during the siege led to the unconditional reduction of the inhabitants to captivity in a way that, when they finally surrendered, they were completely deprived of any right, totally at the mercy of the sovereign’s will by the laws of war.²⁶

24Mateo Páez pointed out two other aspects to differentiate between slaves and captives. First, he insisted on the slave’s profitability as merchandise, as opposed to captives. This scholar attributed a work-related or speculative use to the slave, considered as labour force, or a commercial luxury article, provided by international trade. By contrast, the captive always had the doors of redemption open.²⁷

25To my mind, this statement requires nuance. As for the possibility to regain freedom, an issue on which Professor Hinojosa has also insisted, I totally agree.²⁸ But I do not ascribe work-related use exclusively to slaves, for captives could also be used in this sense. Furthermore, we have seen that the Partidas recognised that those who had captives in their hands “use them as serfs”.²⁹ This notwithstanding, labour use does not modify the very nature of the concepts, for the captive, no matter if he worked for his captor while waiting to be rescued, did not cease to be in a provisional and temporary situation that did not necessarily ended in slavery.³⁰

26Likewise, the status of merchandise can be ascribed to both the slave and the captive, for captivity nourished a flourishing commerce both in medieval and modern times.³¹ More concretely, North African captives were the most valuable booty for the people that repopulated the Kingdom of Granada after the Castilian conquest, and from the other shore Muslims could demand several goods, particularly silk, to rescue Christian captives.³² In fact, Christian captives were highly valued, frequently exchangeable by products of limited distribution. This circumstance prompts me to point to economic value as another difference between captivity and slavery: from captives, captors expected a considerable ransom, whereas from slaves owners sought mainly their working capacity.³³

27We must take into account another point that differentiated captivity and slavery: the temporary nature attached to the former. As a matter of fact, the possibility of recovering freedom after paying a ransom was real for captives, while slaves depended not only on their own economic power but, more importantly, on their owners’ will, beyond the right legally spelled out to attain liberty. Professor Furió underlined this aspect when he stated: “captivity is a temporary and discontinuous phenomenon, the result of the frontier’s mobility and permeability, which allowed both Christians and Muslims to capture enemies of their faith in order to be paid a ransom”.³⁴

28Professor Furió’s assertion is valid not only for the terrestrial border but also for the maritime frontier, the Alboran Sea that divided Iberian Christendom from the Islamic Maghreb. Not all the Muslims captured in North Africa became slaves. Some, not many to tell the truth, retained their condition of captives at the proviso of paying a ransom.³⁵ At the beginning of the summer of 1498 Juan de Lezcano, captain of the Royal Navy, lead a hostile raid in the surroundings of Badis (Vélez de la Gomera in the Spanish documents), which brought a return of 24 captives, to whose ransom he agreed before the public notary García de Villoslada on 9 July. He agreed to different amounts for each captive, to be generally paid within six months, an arrangement that yielded him 3,550 doblas hacenes (1,599,750 maravedíes), a great sum that explains why Captain Lezcano preferred to ransom his captives rather than to sell them in public auction, even if three of them never paid after escaping.³⁶ Once more, economic value and temporary nature were differentiating features that highlight a distinction between captivity and slavery.

Muslim captives that became slaves. Some examples from the Kingdom of Granada

29We do not know all of the circumstances surrounding the ransom of the people of Malaga, who were reduced to captivity after the surrender of the city in August 1487. But I want to emphasise the fact that they were immediately given the possibility to recover their freedom by each paying a ransom of 30 doblas hacenes, an option of which some Muslims and the whole Jewish community took advantage.³⁷ Yet, in practice, the high cost of the ransom made this option impossible for the overwhelming majority of the Islamic population of Malaga. The direct consequence was that they were delivered to communities and individuals “more as slaves than as captives”.³⁸ Therefore, we may conclude that in the Crown of Castile the step from captivity to slavery was determined by the economic ability of the captives to assume their own ransom, a matter already considered by Visigoth laws in the seventh century.³⁹

30Following this reasoning, we have to consider the heterogeneous character among the captives from Malaga. As Professor Ladero has pointed out, first we have to distinguish between Muslim captives destined to be exchanged for Christian captives, and the rest of the captives, kept by the Crown, who numbered about 8,000 people. Of these, from 2,500 to 3,000 were distributed by chance among the military chiefs of the army, high and low nobles. But, as the chroniclers Bernaldez and Valera put it, they had not been inhabitants of the city, but refugees who had gone to the city before the siege and the gomeres, the Maghreb volunteers who had arrived to help the resistance. Furthermore, according to Valera, they were not even offered the possibility of ransom.⁴⁰ The rest were sent mainly to Seville and Cordoba, but also to Valencia and Barcelona, in the Crown of Aragon, where most of them were sold at auction as slaves in local markets.⁴¹ Their status as slaves is later confirmed when we know that some of them, mainly from Seville and Cordoba, escaped from their masters and attempted to reach Malaga, the door to the Maghreb and freedom, or Granada, the capital of the kingdom, seeking to take refuge behind the walls of the city, where they could live free.⁴²

31The religious proselytism practised by cardinal Cisneros and the failure to fulfil the capitulation agreements led to the Mudejar rebellion in 1500, with irreparable consequences for the Muslim population of the kingdom. Those who had revolted against the Crown were once more reduced to captivity. The so-called “hostages of Daydin” offer the most demonstrative example of Muslim captives becoming slaves for not paying the ransom decreed by the Catholic Kings. In fact, once the revolt was extinguished in 1501, the Mudejars from Daidin, an alquería (farmstead) north of Marbella, were sentenced to perpetual exile, the loss of their possessions and the collective payment of 10,000 ducats (3,750,000 maravedíes), in pledge of which some captives were delivered, a circumstance which has rightly assimilated them to the captives of Malaga. It seems clear that the ransom was not paid, for in May and June 1502 the hostages were sold as slaves in public auctions in the city of Malaga. Once more, the non-payment of the ransom resulted in the captives becoming slaves. Yet, in this case there is a new feature that differs from the circumstances of the captives of Malaga: the collective ransom was very high and difficult to pay, but the Morisco community could more easily afford the individual purchases of the auctions, even at the risk of not being able to buy all the captives to legally give them back their freedom, as it happened.⁴³

32Mudejars from all over the Kingdom shared the same fate. In the east, in the bishopric of Almeria, the Mudejars of Inox, Huebro, Tarbal, Turrillas and Nijar also revolted against the Crown and for the same reasons. All the captives were given the possibility of paying a ransom, except those from Nijar, who had already rebelled against the Catholic Kings in 1488, less than a year after the Christian conquest, during the War of Granada. The monarchs did not forgive the second uprising and rejected the proposal of alfaquíes (Muslim doctors), local wazirs and “good men” to pay 25,000 ducats (9,375,000 maravedíes) in three instalments within 20 months for 790 inhabitants of Nijar who were directly enslaved by the laws of war, without having been considered captives. Some captives from the other villages could not pay their ransom, and in spring 1501 slaves from Nijar and Huebro were sold in public auction for 4,528,000 maravedíes. Some had fled from their captors in Nijar, some while being transferred to Cordoba. Others from Huejar and Lanjaron, in the Alpujarras, did not pay their ransom and were sold as slaves to new masters from Granada, who shortly after having purchased them in auction reported to the kings that the slaves had fled from them with the help of friends and relatives, and returned to their homes, where they remained in hiding.⁴⁴

33The Spanish conquest of North African cities in 1508-10 – most notably Oran, Algiers, Tripoli and, in 1535 Tunis – led to the enslavement of the Muslim population. The quinto real for the slaves of Oran was collected in Malaga in a thorough operation organised by the Crown.⁴⁵ The population had been enslaved following the laws of war, without having been considered captives with the right to pay a ransom.

34Once more, we find that in all the cases presented Muslim captives were offered the possibility of paying a ransom to recover their freedom, and when it was not fulfilled, they became slaves. In the case of the Mudejars from Nijar or the Muslims from Oran they were directly enslaved. Therefore, we confirm with other examples the economic and temporary features that separated captivity from slavery, and the condition of the former was a previous step before the latter.

Islamic law and the Maliki school: Why Christian captives were not slaves under Muslim masters

35Up to this point I have developed two lines of argument to emphasise the differentiation between captivity and slavery. The first one addresses the ideological, legal, economic and temporary features. The second underlines that captivity preceded slavery, a fact that is clear in the proceedings from the Crown of Aragon,⁴⁶ and was established by practice – the fulfilment of a ransom or not – in the Crown of Castile, including the Kingdom of Granada after the Christian conquest. But there is a last question to deal with, which scholars claiming that Christian captives in North Africa were actually nothing less than Christian slaves with Muslim masters always omit in their reasoning: Islamic law and the Maliki school applicable to western Islamic slavery.

36As a recent historiographical tradition clearly shows, although there are some analogies and coincidences in the situation of captives and slaves, and they even overlap sometimes, the juridical status of slaves under Islamic law widely differed from the captive’s status, and the law treated them separately. It is a juridical difference that becomes clear following the Qur’an, the Sunna (Traditions and Practices of the Prophet) and the unanimity of classical jurists.

37The terms to define Muslim captives and Christian captives in Arabic, in the Qu’ran and Islamic law in Nasrid Granada, were different, and they were clearly differentiated from those used to name slaves. Even though the expressions to refer to slaves imply different nuances, sometimes complex to determine, it is absolutely clear that they are never confused with the words used for captives.⁴⁷

38Furthermore, the fate of prisoners of war totally depended on the will of the Imam, who at this time was the political leader, independent of his title (caliph, emir, sultan). Once they had lost their freedom, captives faced four possibilities: amnesty, ransom, execution or slavery.⁴⁸ Therefore, in Islamic law captivity was also a prior step to slavery, but did not necessarily lead to it.

39In the transition from the Middle Ages to modernity captives were seldom executed, and only Muslim sovereigns could grant amnesty, usually as a gesture of goodwill. Slavery was a condition widely reserved for black Africans, even when they were Muslims and despite Islamic law, which formally banished it, as the Miʿrāj al-ṣuʿūd, a fatwa or legal opinion written in Timbuktu by Aḥmad Bābā at the end of the sixteenth century, shows.⁴⁹ Therefore, our attention will focus on ransom dispositions.⁵⁰

40We cannot proceed without taking into account that in al-Andalus (Nasrid Kingdom of Granada) and the Maghreb the Maliki school ruled. Professor Charouiti Hasnaoui has carefully analysed the fatwas (legal opinions) by two fifteenth-century Muslim jurisconsults of Malaga, al-Wansharisi and Ibn Tarkat, which mostly examine the casuistry of captivity.⁵¹ The legal consultants established four categories of captives: elder men, minor children, weak men (ancient and sick), and women. In theory, elder men could only be released by the sovereign, or by private individuals if in exchange for a Muslim captive. The rest could be released after paying a ransom, although the case of minors was very controversial. While most of the jurists of the Maliki school were opposed to ransom, practice and the need to finance defence against Christian hostilities prevailed. From captives, captors mainly expected a profit, as the dynamic in the Alboran Sea reveals at the end of the fifteenth century and for the first third of the sixteenth century.⁵² It is clear that they were not considered slaves. And even if it is true that in the particular case of the Barbary Regencies captives who could not afford a ransom were often destined to galleys as rowers, their treatment and consideration differed from that reserved for black Africans. Without a doubt, Christian captives were not slaves with Muslim masters, either from the Christian point of view, or from the Islamic one.

41To summarise, slavery and captivity differed in the historical genesis of the concepts and, consequently, in the procedures to recover freedom. Besides, there were ideological, legal, temporary and economic divergences. In the 15^th century western Mediterranean the steps to evolve from captivity to slavery were essentially clear both from the Christian and Muslim points of view. Therefore, analogies and coincidences pointed out by scholars who have mainly considered an economic perspective (trade, working capacity) should by no means lead to a confusion of the very nature of both conditions. Consequently, Christians should always be referred to as captives under Muslim captors; black Africans always as slaves; and Muslims (including “black Moors”) as captives or slaves depending on their particular situation.