Infirmities and invisible ink: enslaved Muslims and magic in Malta, c.1598–c.1608

This article examines accusations of magic made against enslaved Muslims in Malta over the period 1598–1608. In this period numerous enslaved Muslims were accused of magic before the Roman Inquisition, and the surviving records describe these accusations and magical practices in detail. This article explores the types of magic Muslims were accused of, and why they seem to have been attractive to Christians as magical practitioners. It employs a statistical analysis to outline patterns of accusation, and uses the detail from individual cases to discuss these patterns. The authors argue that accusations against enslaved Muslims had several distinctive features. Muslims were believed to be particularly good at curing illnesses caused by sorcery, which was linked to wider early modern stereotypes about Islam. They were also more likely to use books and writing in the Arabic language. Finally, their practice of love magic for Christian women crossed gender boundaries in a way that male Christian practitioners did not, although this particular aspect should be seen in the context of a wide variety of types of magic practised for both men and women. However, despite these differences, their services were not seen as completely distinct from those of Christian practitioners. Magic was thus an area where people of different faiths interacted, especially in certain circumstances. This picture corresponds to some degree to patterns identified in recent studies of enslaved and non-Christian healers who came before the Inquisition in Latin America and Africa, and the article thus offers suggestions for comparison with these studies.

these places Muslims might also come before the Inquisition accused of magic.Magic accusations against enslaved Muslims have received comparatively little attention from scholars researching Spain, Sicily, or Italy, but Cesare Santus has discussed a case from Livorno in detail, and Stephen Haliczer noted a reference to a Muslim diviner in Spain. 11owever, Malta was probably the European country with the highest proportion of accusations in relation to its population. 12Precise figures are hard to obtain, but estimates based on evidence from 1610 and 1632 suggest there were perhaps around 2000 Muslims on the Maltese islands (around 1500 belonging to the Order of St John, and an unknown number in private households), constituting around 3.5% of the population.They were largely concentrated in the harbour cities of Valletta, L-Isla (Senglea), and Birgu (Vittoriosa) (Figures 2 and 3), such that in those areas up to one in four of the male population may have been enslaved, therefore making up a very substantial minority. 13ost Muslims on Malta were North Africans, referred to in the Inquisition records as "Moors" (mori): we hear of slaves from Egypt and Djerba (Tunisia), for example.There were also a small number of people from the Levant and the Ottoman Empire, referred to as "Turks" (turchi). 14Many were conscripted as oarsmen on the Order's galleys, but others were employed on land, typically building fortifications, or working in private households.Many, but not all, were men: women made up perhaps a tenth of the number, mostly working in domestic households. 15hile on land the men resided in the slaves' prison in Valletta, and were free to move around during the day, earning money in the nearby cities and countryside from activities such as selling water and oil, or acting as barbers; furthermore, their trading activities brought them into frequent contact with the Christian population. 16n front of the Inquisition, Christian witnesses spoke of meeting "Moors" on the road, or at the gates of Valletta, or in shops.17 This level of casual interaction, described in detail by Godfrey Wettinger, was a source of concern to the authorities on the islands, who worried especially that it could lead to sexual relationships between Muslim men and Christian women.The diocesan synods of 1610 and 1625 denounced this as a reserved sin, which could not be absolved by the parish priest.The rulers of the islands, the Order of St John, also forbade these relationships: a proclamation from 1658 prohibited Muslims and Jews from entering Christian houses because of the risk of sexual mixing, and forbade them from interacting with prostitutes in particular.18 As well as sexual mixing, the Inquisition was also concerned about interactions connected with magic and, as we have seen, previous scholars have noted the high number of magic accusations made against Muslims.Nonetheless, we still know comparatively little about what actual kinds of magic Muslims were supposed to have performed, or why they may have proved attractive to Christians as magical practitioners.Did Christians seek out Muslim practitioners for particular reasons, or to deal with certain problems, or were Muslims simply more vulnerable than Christians to being accused before the Inquisition?More broadly, how did magic, and the regular contact it implied, interact with perceptions of Muslims as "Other"? Ths article will investigate these questions, extending work by Carmel Cassar, Frans Ciappara, Godfrey Wettinger, and Anne Brogini, as well as the recent edition of the trial of Sellem bin al-Sheikh Mansur.19 It will argue that imputations made against Muslims had several distinctive features, which suggest that the accused offered particular services to Christians (or were believed to do so), rather than simply being more vulnerable to accusation.One was a perceived specialism in diagnosing and curing illnesses that were thought to have been caused by sorcery.The second was the use of ritual practices that involved the Arabic language and writing; and the third was an ability to cross gender boundaries and offer love remedies to Christian women.Nonetheless, these distinctions are not clear cut: evidence shows that Christians offered similar services, and so there was also a degree of shared ground between Christian and Muslim practitioners.
In exploring these questions this article also aims to speak to the growing body of studies of magical healing and the Inquisition in the early modern Caribbean, Latin America, and Mexico.One focus of this work has been to show that ritual healing was a sphere in which practitioners and clients from a wide range of ethnic and religious backgrounds interacted and shared knowledge. 20Enslaved healers participated in these exchanges, and for them medical practice could offer a route to earn money and gain greater freedom. 21Malta allows us to explore the ways in which similar processes existed in Mediterranean Europe.In Malta, enslaved practitioners likewise commonly interacted with the Christian population, providing a wide range of magical services which drew (at least in part) on Islamicate traditions, but proved attractive to Christians.

Sources: the Roman Inquisition and its records
The Cathedral Archives of Mdina, Malta, house one of the most complete collections of the Roman Inquisition Tribunal, running into 172 volumes with around 30,000 cases from 1561 to 1798 (referenced in the archive as AIM Proc.Crim.).They vary in length from simple denunciations of a single page to detailed proceedings and investigations that might run to a hundred or more folios.A small number of these cases led to a formal trial, or "processus criminalis", but most were court appearances, labelled "denuntio" ("denunciation") or "informatio" ("information") whereby witnesses who had been involved in magic practices came to the Inquisition to confess their offence.Thus relieved, they were pardoned and sent home, and no further action was taken.This, at least, was the procedure for Catholic Christians.On Malta, Muslims were free to practise their religion, and some possessed a copy of the Qur'ān.The Inquisition was not concerned about their practice of Islam, but they could be prosecuted when they were deemed to have led Christians astray.Prominent among these cases was when they were accused of practising magic.
Inquisition records provide detailed descriptions of numerous "magical" rituals -for healing, love formulas, and various other purposes -but they do not offer a transparent window onto these practices.They only include cases that people reported to the inquisitors, and we have no way of knowing how many instances of magic went unrecorded, or how representative the surviving cases are.Moreover, although people from all levels of society appear in the records, certain groups -such as Muslims and women described as prostitutes -may have been especially likely to be denounced, as discussed below.The circumstances in which people made denunciations are also important.Most confessions are described as "spontaneous", but this is misleading.Magic was one of the sins that was reserved to the Inquisition, and so if penitents mentioned it during confession then the priest was under moral obligation to send the penitent to the Inquisition before granting absolution, and many "spontaneous" witnesses came forward for this reason. 22This message was reinforced by numerous edicts promulgated by the inquisitors.For example Fabrizio Verallo (inquisitor 1600-1605), shortly after assuming the role, published an edict in January 1601 ordering anyone who knew about magic to come forward and tell the Inquisition. 23Other witnesses came forward in order to pre-empt being denounced by someone else.Testimonies made to the Inquisition were meant to be kept secret, but in a relatively small community it was possible to guess if one was at risk of being denounced. 24nce at the inquisitor's palace, witnesses' testimonies were shaped by many different forces.Most importantly, they were led by how the inquisitors categorized magic.The terms most often used in the records for magic were maleficium or sortilegium in Latin, or magaria in Italian.Rituals or objects were also referred to as "remedies": remedia in Latin, rimedia in Italian.Maleficium, sortilegium, magaria, and remedia/rimedia covered activities and objects used for many purposes, including healing, harming, countering witchcraft, and love magic, and although maleficium carried an implication of using magic to cause harm, the documents often do not draw sharp distinctions between these terms.Sellem bin al-Sheikh Mansur, for example, was said by the Inquisition (in their Latin summary of his case) to have performed maleficia (magical remedies) to cure other maleficia (harmful spells), and to have carried out "maleficia and sortilegia" for love. 25ther terms were used for more specific activities, such as nigromantia for rituals that invoked demons explicitly.For inquisitors, what made all these activities magic was that they were not thought to work by natural means, and so must involve demons, whether or not those demons were openly invoked.Here the inquisitors were following standard theological views of magic developed in the Middle Ages, as set out in inquisitorial manuals such as Nicholas Eymeric's 1376 Directorium Inquisitorum which was reprinted and influential throughout the early modern period. 26They were especially concerned with magic when it involved heresy, that is, incorrect belief: for example if someone knowingly invoked demons, believed demons could do things that were beyond their power (such as forcing someone to love or hate another person, since the Christian doctrine of free will held that this was impossible), or misused the sacraments. 27One form of magical heresy did not feature very often, however: like many other inquisitorial bodies (Spanish and Roman), the Roman Inquisition on Malta took little interest in the sort of diabolical witchcraft that led to witch trials in northern Europe, and inquisitors were cautious about claims that someone was part of a secret, devil-worshipping sect that used magic to harm others.While inquisitors on Malta did pursue a few accusations of trafficking with demons (for example, the case of Betta Caloiro, discussed by Carmel Cassar), inquisitors often treated these cases with scepticism and relative leniency. 28hese inquisitorial priorities shaped the questions witnesses were asked and the responses they gave.For example, in order to determine whether someone had committed heresy, questioning often focused on what the witnesses had believed they were doing.Were they aware, or did they suspect, that a certain remedy was forbidden, or involved demons?Did they believe a love remedy could change a person's mind from hate to love, or vice versa?In response to these leading questions, witnesses often pleaded ignorance: they said they were not sure whether a ritual would work, or how it worked, or they did not realize it was forbidden.Responses such as these might reflect genuine ignorance on the part of the witnesses, but they appear so often that it may have been easy to guess that this was a way to secure a lenient response from the inquisitor.They may also reflect coaching by the notaries at the Inquisition tribunal or (in the case of Christian witnesses) by their confessors and parish priests. 29 further challenge in using Inquisition records is that they do not always preserve the witnesses' exact words.Notaries were supposed to write down everything that was said verbatim, but in practice this was complicated, especially in Malta where multiple languages were spoken.Testimonies were written down in Tuscan Italian, but when witnesses spoke in Maltese the notary translated their words.Notaries generally seem to have recorded witnesses' words closely, sometimes preserving Maltese terms they used, but we cannot be sure of this in every case. 30Common phrases recur, such as witnesses saying they have come "to clear their conscience" or they "cannot imagine" why they have been called.These may reflect paraphrasing by the notary, or again coaching by a notary or confessor in how to present one's activities before the inquisitor. 31onetheless, despite these challenges the records tell us about expectations and beliefs relating to magic, even if they cannot reliably show exactly what a certain individual believed or did.They are useful for examining patterns of accusation -who was accused of what -and for descriptions of magical practices that are likely to have seemed plausible to the witnesses and inquisitors.This is all helpful for understanding what services Muslims were believed to offer, and why they were attractive as magical practitioners.It also seems that in at least some cases, the descriptions of magical practices were accurate.As we will see, in a few cases the testimonies are corroborated by magical objects and papers preserved in the records, with these surviving objects matching the descriptions in the text.

Methods
To examine the pattern of accusations against Muslims we used a database to record cases of sortilegium, maleficium, magaria, rimedia/remedia, and nigromantia.This was put together from two pre-existing databases, compiled independently: a database of magic cases compiled by Catherine Rider and Dionisius Agius in 2019-2022, covering the years 1598-1608, 32 and a database of all Inquisition cases compiled by Gabriel Farrugia in 2020-2022, covering the years 1600-1620. 33The three researchers combined their results for the period 1598-1608 which corresponds to the inquisitors Antonio Ortensio (1598-1600), Fabrizio Verallo (1600-1605), Ettore Diotallevi (1605-1607), and Leonetto Della Corbara (1607-1608).The cases are not kept strictly in date order, so the database includes a few outliers that began at earlier or later dates, and may omit a few cases.Nonetheless it includes the vast majority of cases from the period.
We chose this 10-year period because as noted above it was one when the Roman Inquisition, on Malta and elsewhere, was dealing with high numbers of magic cases.Because case numbers remained high across the century in Malta, the early seventeenth century was not unusual in this respect, but when choosing a sample period at the start of the project, it made sense to select one in which we were sure we would find large numbers of records.Statistical analyses of inquisition records have been conducted before, including for Malta.Anne Brogini included a general breakdown of accusations made to the inquisition from 1577-1670, while for magic Carmel Cassar analysed the cases from 1646 to 1649, and Frans Ciappara those of the period 1777-1798. 34We are using similar methods, but have focused on cases of Muslims accused of magic, identifying patterns of similarity and difference with what is known about accusations made against Christians.
The information recorded by the database is listed in the Appendix.It includes the accusations made, the names of the accused and any other practitioners mentioned, the dates, and other information about the accused, such as their gender.We also note whether the case involves Muslims, either as the main accused or as another practitioner named in the record.We also recorded the names and genders of Christians who employed Muslim practitioners.For convenience, we have referred to these Christians as "clients", but that term may place too much emphasis on the commercial nature of the relationship.Many Muslims were paid for their magical services, but magic could also form part of informal everyday exchanges between people living in close proximity.

Services: countering and performing harmful magic
In total, we found 219 cases of magic for the period 1598-1608.The vast majority are denunciations or pieces of information, while a few are actual trials.Table 1 shows the number and types of magic accusations made to the Inquisition, and indicates how many of these accusations included references to Muslim practitioners.The total number of accusations (304) is higher than the number of cases (219) because many cases contained multiple accusations.For example a single practitioner might be accused of performing both love magic and healing, or of diagnosing and curing illnesses that were deemed to be caused by magic, and also offering cures for non-magical illnesses.The "Other" category comprises two sorts of cases: those where the purpose of the magic is unclear, such as when a woman was seen throwing items into the harbour and a witness denounced this to the Inquisition as suspected magic (Vol.18, ff.123-24); and accusations that only appear once or twice, such as magic to win a court case (Vol.26C, ff.948-53); magic for contraception (Vol.21A, ff.205-08); or a magic spell for a woman to remain beautiful (Vol.23A, ff.353-56).
Table 1 shows that Muslims were accused of offering many of the same services as Christians.Both groups performed love magic and magical healing in large numbers.Among the Christians, the types of people who performed love and healing magic varied and some of these Christian practitioners had no clear parallels among the Muslims accused.For example the records include relatively frequent accusations against Christian women in informal relationships (sometimes described as prostitutes) of using magic to win or keep the love of their partners, a pattern also found in inquisition records from Italy and Spain. 35Accusations of using love magic on their own behalf in this way are rarely made against the mostly male Muslim practitioners -although as we will see they were accused of love magic for Christian women seeking to retain their partners.However, there were also Christian practitioners who offered magical services to clients in similar ways to the Muslim practitioners described below.Francisco Moneglia was known for healing and producing remedies for lovers (in the same way as male Muslim practitioners did), while a Greek woman named Angelina Rodiota, also known as Pandigliudena, was mentioned in three cases between 1593 and 1599 as someone who created remedies, treated illnesses caused by magic, and countered love magic. 36Muslims were not, therefore, unique in offering love magic or healing, and there were Christian practitioners who catered to similar needs.The table confirms the general picture suggested by other scholars that Muslim practitioners were substantially over-represented in all types of accusation, compared to their numbers in the population (or even their numbers in the harbour cities).In part this may have been because as outsiders in Christian Malta they were vulnerable to accusations.This is difficult to prove, because it is hard to know how many Christian (or Muslim) practitioners were not accused, but it seems plausible.Scholars working on inquisition records elsewhere have likewise argued that foreigners or outsiders were more likely to be accused of magic.Ruth Martin has noted that in Venice Greeks, Friulians (from outside Venice), and Slavs were especially likely to be accused. 37imilarly María Tausiet, working on the Spanish city of Saragossa, has argued that many accusations were made against people from outside the city. 38Studies of inquisition records outside Europe have also suggested that non-Christians were particularly vulnerable to accusations, and could easily lose what local support they had, meaning that clients could become accusers when a case came to court. 39Viewed in this light, it may not be a coincidence that Angelina Rodiota, who came before the Inquisition several times, was Greek.Thus, in part, the numbers in Table 1 are likely to reflect the vulnerability of enslaved Muslims, rather than their propensity to engage in magic.On the other hand, it also seems likely that some Muslims on Malta did practise magic as a way of earning money, perhaps more regularly than Christians did, as Sellem observed in his trial.This again is a pattern scholars have found in inquisition records elsewhere: Tausiet has argued for Saragossa that although outsiders were vulnerable to accusations, magic was also a source of income, which could be attractive to outsiders who had recently arrived in the city and were seeking economic stability. 40A similar pattern may well have been at work on Malta.
Table 1 also shows that while Muslims were associated with magic generally, they were also strongly associated with certain types of magic.They appear in especially high proportions for treasure hunting (86% of accusations) and diagnosing and curing illnesses that were deemed to have been caused by magic (78%).This points to a complex picture, in which enslaved Muslims were believed by Christians to have skills which made them attractive as magical practitioners in certain situations -rather than simply that they were susceptible to all types of magic imputations.
The total number of treasure-hunting cases is small, so it is difficult to generalize from them.It is possible that the high number of Muslims mentioned reflects Christian stereotypes that Muslims might have special access to the demonic, as discussed in more detail below.Hidden treasure was often believed to be guarded by spirits or demons, and so treasure-hunting rituals required a practitioner who could deal with these beings. 41In many parts of Europe dealing with spirits involved Christian prayers or rites, as well as ceremonial magic, and on Malta non-Christian magical practitioners may well have offered an alternative approach to the same problem. 42n the case of diagnosing and curing illnesses caused by magic, we have a larger body of cases to examine, and the high number of Muslims, and the low number of Christians, mentioned in these cases is striking when compared to accusations of magical healing more broadly.A number of inter-related factors are likely to lie behind these figures.On the one hand, they may reflect more general early modern perceptions of Islam and Muslims which viewed them as negative, "other", or even demonic.For example Eric Dursteler uses inquisitorial proceedings to argue that there was a general belief that "the Turk" was a violent and malicious individual, which allowed those who had suspicious dealings with enslaved Muslims, to shift responsibility onto "the other". 43On Malta, where the Christian inhabitants shared spaces with Muslims, artistic representations reinforced these stereotypes in sculpture and paintings in churches, and decorative carpentry on board the galleys.One would have seen on the galleys of the Order the holes for waterspouts shaped as Moorish faces with open mouths or turbaned heads sculpted along the masts used to tie ropes around or even on ships, where the large testa di moro connected the crow's nest to the mast. 44These prejudices also encompassed views of Islam, which could be seen as a demonic faith.Early modern popular representations of Islam often portrayed Muslims as idol-worshippers, and some polemics argued that early Islam's "miracles" were in fact magic. 45 study by Cesare Santus of an enslaved Muslim named Yusuf (Ar.Yūsuf), who in 1611 was accused of magic in Livorno, Italy, suggests how these prejudices might shape accusations against Muslim ritual practitioners.Santus argues that because Christian polemics tended to associate Islam with the devil, it would have been a short step for Christians to see Muslims as experts in dealing with the supernatural.46 Several cases in Malta also fit Santus's interpretation.Berto Briffa, explaining why he asked Sellem bin al-Sheikh Mansur to cure his illness, said "because I believed that my problems had been produced by evil and demonic things I judged that this infidel would know how to solve them".47 Similarly Domenico Felici, a guard at the slaves' prison, approached two Muslims to cure him when he suspected his former girlfriend of using love magic on him.He told the inquisitor he thought they would know about these things "by means of some of the secrets they have in their Muslim sect".48 Thus there were a number of reasons why Muslim practitioners might be called on in cases where someone was needed to deal with magic or demons.Linked to this, both the choice of a Muslim healer, and the suspicion that a person was bewitched, were often presented as a last resort for dealing with illnesses.In several cases witnesses (including Berto Briffa) said they employed Muslim practitioners after Christian physicians and other healers had failed.49 In 1598 Antonina Camilleri called in a Muslim named Xirif (Ar.Sharīf) to cure her daughter, who had a swelling ("postema") on her face and had tried cures from physicians without success.50 In 1599 Antonina Vella told her friend Caterina Zarb that when she was ill, she tried cures from doctors and a religious cure from Brother Aurelio of Mellieħa (west Malta), before deciding she must be under a spell.Finally her husband called in a Muslim named Chaidar (Ar.Ḥaydar) who identified the magic.51 It is possible that claiming they used magical cures as a last resort when other cures had failed was a strategy that witnesses used to make themselves appear less culpable before the inquisitors. Noetheless it is also possible that bewitchment was suspected in difficult cases which resisted other forms of healing -and that these were also the cases in which Christians might be more likely to turn to a non-Christian healer after exhausting other options.This pattern of accusation might therefore reflect the link between Muslim practitioners and magically caused illnesses, but also reinforce further in the minds of witnesses and inquisitors that Muslims were the people to deal with sorcery.
These figures do not suggest that dealing with magical illnesses was exclusively a Muslim activity.Nonetheless, Muslim practitioners are markedly over-represented in these accusations.Within this combination of prejudices and opportunities, it is likely that some enslaved healers found a space to act for their own advantage, as also happened in other parts of the world.It was possible for them to do this because some Muslim practitioners had distinct skills and magical practices at their disposal.These will be explored in the next section.

Skills: Arabic language and writing
Descriptions of magic performed by Muslims regularly involved reading, writing, and the use of books or written papers.Of our sample of 219 cases of magic (performed by either Christians or Muslims), 80 involve reading or writing.Of those, 48 (60%) also involve Muslims.These practitioners are described as using papers written in Arabic script, or containing unidentified characters that appeared to Christians to be Arabic, often written in yellow ink.Margarita, described as "Ethiopian" (probably meaning black), was accused of magic when her mistress found a bag containing hair and writings in yellow ink done by "an infidel Moor". 52Mimichella de Patti and her mother Caterina were put on trial for, among other things, owning a magical bag made from green damask, with silver thread, which contained a long, wide paper with Arabic letters written in yellow ink. 53Mimichella confessed that she had been given the paper by a Muslim, who had visited the house to deliver water and found her sad because her husband mistreated her. 54In 1605 Joannes Gavone said an unnamed Muslim had tried to identify a thief for him by writing four papers in yellow ink, on which he wrote the names of the people Gavone suspected. 55Most of the practitioners who wrote magical papers were men, but not all: Hali (Ar.'Alī) tells of a woman, Haixa (Ar.'Ā'isha), writing two Arabic magic papers. 56n some cases textual amulets produced by Muslim practitioners contained series of unconnected letters, a practice attested in medieval and early modern Arabic magical texts. 57Other amulets probably contained Qur'ānic verses for healing or protection, since similar papers were well known in North Africa and elsewhere in the Islamicate world. 58riting holy scripture for protective or healing purposes was not solely a Muslim activity: in both Islamic and Christian contexts in this period, written amulets often mixed "devotional" and "magical" elements in ways that could not be easily distinguished. 59Nonetheless, amulets containing Arabic script would have appeared mystifying to Christian clients who did not understand the script.
A small number of magic papers surviving in the Inquisition's records show that the papers described by witnesses were a reality. 60For example, one textual amulet comes with a pouch and gold thread with faded yellow Arabic writing on paper which is now worm-eaten (Figures 4 and 5).Agatha Morena, a former prostitute, said a Muslim had given it to her five years previously and promised it would keep her beautiful. 61nother Arabic magic paper owned by a Christian healer, Imperia Galea of Mosta, consists of 51 boxes, each box with three letters; the paper was strung when folded and was used as an amulet to be worn around the neck (Figure 6). 62Squares or grids containing letters or numbers are numerous in talismans from the premodern Islamicate world: these sometimes represent letters from Qur'ānic verses, or Qu'rānic letters translated into numbers. 63Because they were so common it is not surprising to find Muslims making them on Malta.
Sometimes the records refer to books written in Arabic, rather than simply papers.In 1598 Xirif Habdil Chasem (Ar.Sharīf c Abd al-Qāsim) and another unnamed Muslim were said to have read from a book and recited invocations as part of a ritual in which they diagnosed Marietta, the wife of Francisco Ferrer, as bewitched. 64In the same year  Domenico Felici confessed that he employed Brahim (Ar.Ibrāhīm) and Sellem (Ar.Sālim) to counter a love spell he suspected had been cast on him by a former lover.Sellem used a libro morisco, a book written in Arabic. 65In 1604 another practitioner named Salem -probably Sellem bin al-Sheikh Mansur -was accused of owning a book in which he "divined" that an illness was caused by magic. 66It is not clear what these books were, but the accused often told inquisitors they were the Qur'ān, which they were permitted to own.This was plausible, since Qur'ānic verses could be used for healing, but it may also have been a way to persuade the inquisitors that their books were legitimate religious works, rather than forbidden magical texts.
The language of these Arabic books and papers was instrumental to magic practice.Both Latin and Arabic magical papers were ways to communicate with supernatural powers, relying on powerful words, names, and characters, but Arabic in particular was often viewed by Christians as a language of power that could communicate with the spirit world and cure the sick or help the lovelorn.For this reason, Arabic papers were seen by the Inquisition as strong evidence of magic.A particular concern for the Inquisition was that the Arabic script could contain the names of demons or spirits.Maltese people outside the Inquisition, at a variety of social levels, seem to have shared this belief.For example, Dionisio Cardona, a carpenter, said he believed the "Moors" could make remedies with the power of demons. 67That some of the written Arabic on these scraps of paper could contain Qur'ānic verses did not matter; it caused as much concern to the inquisitors as other wording of profane content. 68nterestingly though, no Arabic word for "demon" appears in the papers examined by us.For the inquisitors, who followed mainstream Christian theology, demons were invariably evil, but Muslim practitioners may have seen things differently.Muslims believe in jinn, "fire spirits" which are sanctioned in the Qur'ān.They can be good or evil and can appear in the form of a human, animal, or some other thing (such as lamps, bottles, and rings). 69The name jinn never occurs in the inquisitorial records under investigation, but it is possible that some of the Muslim practitioners who are said to have invoked demons were in fact dealing with jinn.
Enslaved Muslims are therefore described using books, papers, and writing regularly as part of their magical practice in a way that was less common among Christian practitioners.This is not to say there was a clear-cut division between literate Muslims on the one hand, and illiterate Christians on the other.Many Muslims were probably illiterate, whereas some Christians used books and writing.In 1605 Vittorio Cassar told how a contact in Messina had sent him three magical texts, and asked for his help in interpreting them. 70Gabriele de Piazza, an observant Franciscan friar, owned a small book of "secrets" for divination. 71Some Christians also owned amulets written in European languages: for example Joseph Moneglia told how his brother gave him a piece of paper containing "prayers" to protect him in war, while Isabella was said to have owned "papers" for love magic. 72onetheless Muslims on Malta seem to have offered a more distinctively written form of magic, in a different language, than Christian ritual healers, drawing on written magical and religious traditions from their homelands.Only a small minority of Muslims would have attended elementary schools, but those who had done so would have had some knowledge of reading and writing classical Arabic, which consisted of reading and reciting the Qur'ān. 73A few practitioners, such as Sellem bin al-Sheikh Mansur (who practised geomancy), were familiar with learned types of magic.However, being able to write in Arabic for magical purposes did not necessarily mean wider literacy; for some, learning enough Arabic to perform magic was not a profession nor a hobby, but an aid to survival that could be taught independently.Thus we learn in 1602 of Arabic being taught to Muslim slaves for magic practice. 74In another case a Muslim woman named Xiamexa (Ar.Shmaysha) stated that Chogia (Tur.Hoca), a "Turk", taught her words to recite for love magic. 75This is likely to have been a rudimentary Arabic, a language with a specific purpose for use in magic.It might not have been intelligible to Arabic readers who were unfamiliar with magic: in the 1605 trial of Sellem bin al-Sheikh Mansur an Arabic expert, Giovanni de Cuisa, was brought to court to translate the written magic Arabic but could not do so, as he had no expertise in such a formulaic languagealthough he said they resembled other magical amulets he had seen. 76

Clients: everyday interactions with Christian men and women
In the case of both services (with a focus on countering magical illnesses) and skills (with a focus on Arabic language and writing), accusations against enslaved Muslims, in part, reflected their status as non-Christian outsiders who were believed to have access to distinct magical traditions and connections with demons.In that sense they and their magic were "other".The degree of otherness was always flexible, however, because their practices and services overlapped with those offered by Christians.The idea that Muslims were a distinct, "other", group becomes more complex when we examine the Christian Maltese who sought out their services.Here we see a high level of interaction between Christians and Muslims that often paralleled the types of exchange found between Christian practitioners and their clients, but occasionally differed from itnotably in relation to love magic.
The level of interaction between Christians and Muslims in magic cases varied.Often it seems Christians approached Muslim practitioners to perform magic on their behalf, without becoming directly involved in magical practice themselves.For example, suspecting his wife of adultery, Martino Valletta reported that he and his brother approached Chasem (Ar.Qāsim), looking for a remedy to kill the woman. 77In a few cases the interaction was more sustained than this, and Christians worked with Muslim practitioners to perform magic together, or learned magical techniques from them.Vittorio Cassar, a knight of the Order who appeared many times before the Inquisition in connection with magic, claimed to have learned geomancy from Sellem bin al-Sheikh Mansur. 78Francisco de Saavedra, the captain of the galley Santo Aloisio, accused Vincenzo Romano of doing magic to harm him together with a galley slave, Xirif (Ar.Sharīf). 79Another interesting case from just outside the period is that of Nicola and Matteo Debono, who obtained the Grand Master's permission to dig for treasure at Għajn Tuffieħa (west Malta), after Nicola had a dream of finding this treasure.They enlisted the help of two Muslim galley slaves and a Muslim woman who cast spells invoking spirits.A priest, who was alleged to have composed music for the devil, was also involved. 80agic could thus occasionally be an area in which Christians and Muslims worked together, as well as one where Muslims sold services to Christians.This did not mean the Christians involved ceased to perceive Muslims as other, or as uniquely placed to deal with the demonic.However, close contact had the potential to lead to more complicated interactions that made it harder for Christians to distance themselves from Muslim practitioners.
One type of interaction that spoke both to Muslims' otherness, and to their proximity to Christian Maltese people, was their practice of magic on behalf of women.Carmel Cassar has argued that this was a distinctive feature of enslaved Muslims' magical practice, noting that they often performed love magic for women, especially prostitutes (corteggiane). 81Enslaved Muslims did indeed do this type of magic for Christian women.For example, in 1602 Salvo Camilleri claimed to have seen two Christian women and a Muslim man preparing a magical remedy for impotence. 82In testimonies from 1605 Hali (Ar.'Alī) was said to have offered love magic to Dominica Cassia: he recited words over her gold earring and fumigated it, and said that anyone who saw the earring would be well disposed to her. 83 However, these cases of love remedies for Christian women are only part of a more complex picture of interactions between Christians and Muslims in the area of magic.Table 2 investigates the services offered by Muslim practitioners and the gender of their clients further.Of the 85 cases in which Muslims were mentioned as doing magic, 46just over half -also involved Christian women as clients, either alone or with Christian men.Muslim practitioners had similar numbers of male and female clients in most categories, except for divination and treasure hunting, which were requested more often by men (though in small numbers).The data is not always easy to interpret, because the gender of the client may mask the involvement of other people in the case.In some healing cases the decision to consult a Muslim practitioner may have been discussed between family members of both sexes, especially when someone was ill.When his wife fell ill, Dominico Borg from Mosta (central Malta) called in a practitioner named Brahim (Ar.Ibrāhīm), and other Muslims.In this case his wife's aunt had said she was bewitched and recommended seeking out a Muslim healer. 84When Chaidar (Ar.Haydar) diagnosed Antonina Vella as bewitched in 1599, one witness stated that Antonina's husband and her two brothers were present; although Antonina herself later said her mother knew about the incident, but her father and brothers did not. 85Similar discussions among family members may have occurred in other cases and gone unrecorded.The evidence that survives thus suggests enslaved Muslims offered a wide range of services to both sexes, going well beyond offering love magic to women.
Even with love magic Muslim men did not have an exclusive relationship with Christian women.Christian men also called on them in 10 accusations, compared with 14 of performing love magic for women.For example, in 1606 Lucas Mifsud procured a love remedy from Hali Gerbino (Ar.'Alī of Djerba) to help his brother marry the woman he desired. 86Furthermore, Christian women also practised magic together, without the involvement of Muslims.In 1602 a certain Regina Mensa confessed to performing magic with the aid of two other women, Speranza and Paulina, to ensure the return of her lover. 87Elsewhere, one Francischella, described as a "courtesan", claimed that another woman, Gionannella, gave her a magnet and papers written in Arabic to keep her lover faithful -evidence of how Arabic textual amulets might circulate among Christians even when Muslims were not involved. 88Christian women might also call on Muslim women -rather than men -in the same way as they called on female Christian practitioners.Two sisters, Helionara and Bologna Crispi, were taught by a Muslim woman, Metina (Ar.Matīna), to recite love spells invoking the spirits as well as the sky, moon, and stars. 89In 1606 Jacobus Fiot Farruge testified that Margarita de Stefano, a prostitute, and her nine-or ten-year-old daughter performed magic with the help of two female "infidels", the aim of which was to make Margarita's clients fall passionately in love with her. 90 However, in one respect the pattern for love magic suggests that Muslim practitioners crossed a gender boundary that Christian practitioners did not.We found only one case where a Christian man offered love magic to a Christian woman, without there being female practitioners also involved.This was when Battista Inguanes told Victoria that her abusive husband had a lover, and offered her a remedy to "free" him from this extramarital relationship. 91Thus, if a woman sought love magic from a man, this person was likely to have been a Muslim rather than a Christian.Nonetheless, the cases discussed above show that the relationship between enslaved Muslims, gender, and magic is complex, and Muslim men doing love magic for Christian women is only part of that picture.

Conclusion
Enslaved Muslim practitioners thus offered Christians a variety of magical services, and these were not entirely different from what their Christian counterparts did.Both Christians and Muslims were accused of performing magic for a wide range of purposes, for male and female clients, and they used a variety of techniques to do so, including spoken and written words, and the employment of herbs, and other objects.Nonetheless Muslim practitioners had a specialism -or a perceived specialism -in certain areas of activity.They were called on much more often than Christian practitioners to search for treasure (sometimes alongside Christians), and also to deal with magically caused illnesses (sometimes where Christian "doctors" had failed).Linked to this, they were sent for more often than Christian practitioners to use magic to harm others.They also offered love magic to women, which Christian men rarely did, although this must be seen in the context of a much wider range of services (including love magic) offered to both men and women.
To some degree, these accusations are likely to draw on stereotypes about Islam and Muslims (in particular, stereotypes of Islam as demonic), and reflect the position of Muslims on Malta as non-Christian, enslaved, and potentially vulnerable outsiders.It was not only a case of stereotyping, however.Some enslaved Muslims -although it is hard to be sure how many -had distinctive magical skills to offer, and employed them to earn an income.Their command of an unfamiliar language and script set them apart from Christian practitioners, as did their perceived links with demons.There were, therefore, real differences between the forms of magic that Muslims and Christians were believed to offer, even if there were also overlaps.Kalle Kanonoja, studying eighteenth-century Brazil, has highlighted a similar pattern where black or indigenous healers offered distinctive services to white clients, but were able to appeal to some basic common assumptions, such as a shared belief in the power of words. 92Muslim healers on Malta seem to have struck the same balance: the idea of written amulets and powerful words, for example, would have been familiar to Christians, but Arabic script less so.
In Latin America and Africa, scholars have argued that interaction between practitioners of different faiths went further (at least in some cases), and both Christians and non-Christians blended practices from different cultures. 93There is scope for more research on this in Malta.Detailed studies of individual cases and surviving textual amulets would help us to understand whether, and in what ways, North African magic changed when enslaved healers used it in a Christian environment.However, our impression is that Muslim practices seem to have remained distinct, without any discernible blending of magical traditions.This may be because they often made use of writing and knowledge of the Arabic language, which were harder to merge with the more often unwritten magic of Maltese Christians.It is also possible that Christian practitioners were unwilling to assimilate what they saw as Muslim religious practices: Gunnar W. Knudsen has suggested this for Spain, where Christians seem not to have assimilated Morisco magical practices, which were closely tied to Islam, and the same may be true for Malta. 94More importantly, it may also be that there was little incentive to fuse practices.We have argued that what made enslaved Muslims attractive as magical practitioners was precisely their use of Arabic and their status as non-Christians.Blending practices from different cultures may therefore have reduced their rituals' power (as both they and their clients saw it), and their appeal to Christians.
Thus, despite a few cases of Christians and Muslims working together or teaching magical techniques, most of the time interaction in the field of magic took the form of a transaction, with enslaved Muslims selling their services and performing rituals for Christians, rather than a blending of rituals or collaboration between practitioners from different faiths.This was probably the easiest way for people of both faiths to proceed: it allowed Christians to benefit from "exotic" rituals, and enslaved Muslims to profit from offering them, but it also meant Christians were able to keep a degree of distance.If a ritual went wrong, a Christian client might decide that they had been cheated (as some of Sellem bin al-Sheikh Mansur's clients did), or denounce the healer to the Inquisition.They themselves were not implicated beyond buying a service.
Studying Christian-Muslim interactions in the sphere of magic thus highlights the ways in which practices might be sought and offered across religious and social boundaries, as well as the limits to such interaction.Anxieties about illness, the need for diagnosis and treatment, relationship difficulties, and other common problems, gave people of all social groups an incentive to try multiple solutions, from multiple practitioners, and place less importance on whether a ritual offered by a Muslim (or Christian) practitioner might be potentially demonic -at least until they had to explain their activities to an inquisitor.However, they also show that to an important extent the magic offered (or supposedly provided) by enslaved Muslims remained distinctive.

Figure 4 .
Figure 4. Arabic magic folded and unfolded paper with pouch and gold thread, MCM AIM Proc.Crim.Vol.23A, f. 356r (1605).Reproduced by kind permission of the Metropolitan Chapter Archive, Mdina, Malta.

Figure 5 .
Figure 5. Arabic magic folded and unfolded paper with pouch and gold thread, MCM AIM Proc.Crim.Vol.23A, f. 356r (1605).Reproduced by kind permission of the Metropolitan Chapter Archive, Mdina, Malta.

Table 1 .
Involvement of Muslims in types of magic accusation 1598-1608.

Table 2 .
Magical services said to have been provided by Muslim practitioners to Christian men and women (based on 85 cases which mention Muslims performing magic).