Martyrs and Madonnas: Inácio de Azevedo, the Brazil Martyrs, and the Global Circulation of the Madonna of Santa Maria Maggiore

Abstract

The article offers a revisionist account of the early circulation of copies of the Madonna of Santa Maria Maggiore, known since the nineteenth century as the Salus Populi Romani. Traditionally, the propulsion of the image into global circulation has been attributed variously to Pius V or Francisco Borja, the third Superior General of the Society of Jesus. The article argues that the circulation of the Saint Luke Madonna, as it was known at the time, was closely tied to the martyr’s cult that grew up around the Jesuit missionary Inácio de Azevedo and the so-called Brazil Martyrs, a group of Jesuits murdered by Calvinist corsairs off the Canary Islands in 1570. Azevedo had intended to carry a copy of the Roman icon to Brazil, a copy that perished at sea with Azevedo and the party of Jesuit missionaries. The article suggests that the popularity of the image among Jesuits in Europe and the overseas missions was fueled by the nascent martyr’s cult that followed Azevedo’s death. Painted copies of the Saint Luke Madonna came to function, together with relics of the Eleven Thousand Virgins of Cologne, as proxies for the missing material remains of the martyred Jesuits. The article argues that while the distribution of the image was globally extensive, circulation was restricted to an internal Jesuit martyr’s cult.

Painted copies of the Madonna of Santa Maria Maggiore in Rome began to journey across seas and oceans in the last decades of the sixteenth century. The Saint Luke Madonna, as the image was commonly known, was a well-loved Roman icon (Figure 1). The global circulation of the miracle-working image was an unprecedented phenomenon. While the broad contours of the story are well known, the immediate details of the icon’s diffusion have been allowed to remain shrouded in the mists of hagiographical legend. The habitual account runs along the following lines. The first likeness of the sacred icon was made when the Portuguese Jesuit Inácio de Azevedo traveled to Rome in 1569 to discuss his upcoming mission to Brazil with Francisco Borja, General of the Society of Jesus, and Pope Pius V. Many favours and graces were bestowed upon the missionary. It was Borja who wished Azevedo to have a copy of the Madonna of Santa Maria Maggiore to take to Brazil. Meeting with opposition from the canons of the basilica, Borja appealed to the pope, and it was only through the intervention of Pius V that a copy was obtained. In other renderings, the pope himself gave a copy of the miraculous image to Azevedo. This occurred in May or June of 1569, corresponding to the known dates of Azevedo’s Roman sojourn. When Azevedo left Rome, the story continues, he carried two copies: one for the Brazil mission, the other intended as a gift for Queen Catherine of Portugal, though the copy for Catherine is frequently omitted from the narrative. Azevedo and thirty-nine others were martyred at the hands of French Calvinist corsairs off the coast of the Canary Islands en route to Brazil on 15 July 1570. The copy carried aboard the Santiago, the ship that was to have carried Azevedo to Brazil, was lost at sea. In the wake of this ‘first copy’ of the icon, others followed, reaching the courts of Europe and, via Jesuit intermediaries, the shores of Asia and the Americas.

Like so many layers of Baroque varnish applied to a Mannerist painting, gloss after gloss has been added to this general picture. In the hands of Jesuit historians of the seventeenth and eighteenth centuries, Borja became the first ever to succeed in obtaining a copy of the image, a feat presented as a miracle in its own right. In other accounts, Pius not only refused to allow the image to be lowered from the tabernacle for the copy to be made, but did not permit anyone present while the work was carried out. The anonymous painter who executed the image “had no other witness but the light of the sun”. When the pope saw the verisimilitude of the copy to the original, he became so overcome with emotion that the image was bathed with his tears.1

There are good reasons to question aspects of the story. To begin with, early sources all testify that the first Jesuit copy of the icon was made for Borja’s personal devotional use. We can consider the biography of Borja written by Dionisio Vázquez completed in 1586. Though not unknown to scholars, the manuscript was only recently published and has rarely been consulted directly. Vázquez worked closely with Borja in Rome as Assistant—a kind of administrative under-secretary—for Jesuit affairs in Spain, Portugal, and the overseas missions at the time of Azevedo’s visit to Rome in 1569. Vázquez’s account of the first Jesuit copy of the icon is as follows:

As is well-known, in the Church of Santa Maria Maggiore in Rome, there is an image of the Mother of God, the very same as painted by Saint Luke by his own hand. Having a great desire to have his own true and living likeness (retrato), he [Borja] requested of Cardinal Carlo Borromeo, who had charge of the church, to have a likeness of the image by the hand of a great painter of Rome. And though the canons placed great obstacles in the way of obtaining a likeness of the image, such were the devout prayers and perseverance of the Father that he obtained the image as he wished and placed it in the chapel in the Casa Professa in Rome where he normally said mass.2

There is no reason to doubt Vázquez’s statement that the first copy was made for Borja’s personal spiritual use. Borja’s devotion to the image is well documented. The failure to mention either Pius V or Azevedo, who Vázquez met in Rome, is notable. While Vázquez does not state when the copy was commissioned, other sources indicate that there were at least two Jesuit copies of the image in Rome several months before Azevedo’s visit. One was in Borja’s domestic chapel at the Casa Professa, while the second was at the residence for Jesuit novices at Sant’Andrea al Quirinale.

The testimony of Vázquez is supplemented by the account of Pedro de Ribadeneyra, author of a second biography of Borja published in 1592. Ribadeneyra was a seasoned Jesuit administrator familiar with Jesuit establishments in Rome. He, in fact, replaced Vázquez as Assistant for Spain and Portugal, serving in the post from June 1571 until March 1573 (Maurício 1978, p. 92). According to Ribadeneyra, after obtaining the copy from Santa Maria Maggiore, Borja “placed it in his chapel, and afterwards had other portraits made from it, which he sent to many Princes, nobles, and houses of the Society”.3

None of the early documentary sources indicate that either Borja or Pius V gave Inácio de Azevedo a copy of the icon to transport to Brazil. Certainly, Azevedo carried a painted likeness of the Madonna from Rome. This took the form of a personal devotional gift from Borja to Queen Catherine of Portugal. Borja had developed a spiritual friendship with Catherine during two separate sojourns in Portugal, most recently in 1559–61, when seeking refuge in the kingdom while under investigation by the Spanish Inquisition. Intended for Catherine’s devotional use, Borja requested that it be placed in her domestic chapel. Upon the queen’s death, the image was bequeathed to the Jesuit church at São Roque in Lisbon, where it remains today (Figure 2) (M Borja 1911, vol. 5, pp. 112–13; García Hernán 2004; Henriques 2013).

Early modern embellishment of the narrative of Azevedo and the journey of the ‘first copy’ of the Madonna of Santa Maria Maggiore outside Rome is understandable. By the mid-seventeenth century, canonization campaigns were underway for all three of the story’s protagonists: for the martyred Azevedo, for Borja and, most unlikely of all, for the dour former inquisitor Pius V. While the question of the provenance and circulation of sixteenth-century copies of the icon would reward further study, the purpose of the present article is to re-examine the role of Azevedo and the ill-fated Jesuit missionary expedition to Brazil in launching the icon into global circulation.

The origin story of the first copies of the Madonna has largely gone unexamined. In many modern accounts, Azevedo is left out of the tale altogether, and we are left with only the second, more palatable part of the hagiographical narrative. On this telling, it is variously Borja or Pius V who willed the icon into global circulation, a version of events reified in several recent studies (Mochizuki 2016, pp. 131–33; Mochizuki 2017, pp. 438–42; Mochizuki 2022, pp. 12–17). When Azevedo is mentioned at all, he is portrayed as a sort of first mover, who, having set the first brush in motion, had little to do with the subsequent peregrination of the image. Yet as much recent work has shown, it was devotional cult, not Church hierarchy, that fueled the production and circulation of copies of miraculous images in late medieval and early modern Europe (Garnett and Rosser 2013; Holmes 2013; 2018; Dekoninck 2017; Casper 2021).

If the story of the ‘first copy’ of the Roman icon made for Azevedo is acknowledged to be the stuff of hagiographical legend, we may well ask how and why the narrative was generated in the first place. Rather than dismissing the Azevedo story, in other words, I propose to examine the narrative more closely in the light of documentary evidence that, though not entirely unknown, has not been brought to bear on a critical examination of the circulation of copies of the icon.

As we shall see, sixteenth-century copies of the Roman icon traveled in the company of stories and physical materials associated with Azevedo’s martyrdom. It was only after Azevedo’s fate began to be known that we find traces of the wider circulation of painted copies. Both demand for the image and what little is to be known about the ‘supply’ side of things were the product of an internal Jesuit devotional economy that assigned value to the Roman Madonna primarily due to its associations with Azevedo and the ‘Brazil Martyrs,’ as they were to become known. From the first reports of the disaster, Jesuits learned that Azevedo held the Madonna in his hands as he lay dying. In the dominant narrative, dating from the first weeks following the massacre, the heretical French privateers were unable to remove the image from Azevedo’s hands, which remained stubbornly in the missionary’s grasp even as his lifeless body was thrown into the sea. The little story quickly became entangled with broader currents of confessional violence, Calvinist iconoclasm, and global evangelization.

It is suggested here that within the context of the nascent martyr’s cult that arose in the wake of Azevedo’s death, copies of the Madonna of Santa Maria Maggiore and other objects, including images of Azevedo and the Brazil Martyrs, came to function as proxies for the missing material remains of the murdered Jesuits. Though Azevedo and the Brazil Martyrs were not the first Jesuits to be regarded as martyrs, the spectacularly violent manner of their death at the hands of non-believers, the scale of the massacre, and their identity as missionaries provided an unparalleled focal point for the celebration of Jesuit martyrdom.4

The diffusion of the Madonna of Santa Maria Maggiore is by no means unstudied. It is generally accepted that the image circulated in the company of Jesuits, and there are good accounts of the fortune of the image in the Holy Roman Empire, Ethiopia, India, China, and Japan.5 At the same time, the social and cultural mechanisms that set the icon in motion have been largely overlooked. While in the hands of Jesuit missionaries the diffusion of the image was geographically extensive, the surviving evidence suggests that this was a narrowly circumscribed circuit limited to an internal Jesuit cult.6

All too often, the ephemeral, ‘middle’ stories of early modern mobility are left out of accounts of the migration of people and objects. While early modern mobility has come into closer focus in recent years, the challenge for historians is to place mobility under the microscope, to reject ‘mobility’ as a cypher for distance and geographic displacement, and instead to make localized practices and mechanisms of movement the object of study (Nelles and Salzberg 2023). As scholars have begun to explore how things and people traveled across distances and across cultures, there has been a greater acknowledgement of the ways movement transformed—physically, culturally, and emotionally—objects as they migrated from one location to another. Many things in the early modern world, images among them, were not only mobile—things that experienced the fact of geographic displacement—but were intentionally portable, designed for transit.7 In the case of religious objects, this is not simply a matter of acknowledging the latent agency of material things, but of recognizing the constellation of beliefs, assumptions, and practices that propelled the mobility of things of all kinds. While relics are the most obvious example of this phenomenon, ontological mobility saturated many other types of religious objects, images included.8

1. The Madonna of Saint Luke

The Madonna of Santa Maria Maggiore was one of several miraculous icons in Rome, housed in the city’s first church dedicated to Mary. It is believed to have originated sometime between the sixth and eighth centuries, layered with later medieval accretions and overpainting. It was one of several likenesses of the Madonna attributed to Saint Luke present in the eternal city. Purported to have derived from the hand of one of the apostles, it was regarded as both a contact relic and a miracle-working authoritative likeness. It was venerated for its ability to cure plagues, save children, win wars, and intervene in other ways to deliver the city and its inhabitants from natural and supernatural dangers. It also played a role in the religious rituals of the city. On the Feast of the Assumption (15 August), the sacred likeness of Christ from the Lateran Basilica—also reputed to be from the hand of Luke—was brought to Santa Maria Maggiore to ‘greet’ his mother. In 1613, Paul V installed the icon in the Borghese chapel of the basilica, where it is housed today (Wolf 1990, pp. 31–78; Ostrow 1996, pp. 120–32).

During the years under consideration here, in Rome itself the icon became less mobile and less accessible. Pius V halted the nocturnal procession of the Assumption on the grounds of public order. Though the exact date is unknown, this likely occurred before 1569. In July of that year, Borja informed Catherine of Portugal, in a letter that accompanied the copy of the image carried by Azevedo, that the pope no longer allowed the icon to be lowered from its position in the church, “as he did not wish it to be seen too closely by the people”.9 Though this facet of the icon’s history is not unknown, it urges further caution in taking at face value assertions that Pius V himself supplied a copy to Azevedo.

The naming of the Roman Madonna in the sources is important for understanding the sequence of sacred operations attributed to the icon. It was only in the nineteenth century, as Gerhard Wolf has shown, that the Madonna of Santa Maria Maggiore began to be called the Salus Populi Romani—the ‘saviour’ or ‘protectress’ of the Roman people.10 Early modern sources, particularly the Jesuit sources studied here, universally foreground Saint Luke when referring to the icon. While Rome and the Basilica of Santa Maria Maggiore at times loom large in the telling, in others they are barely mentioned or are occluded. The attribution to Luke, on the other hand, adhered relentlessly to the Madonna on her many voyages.

Luke images were a super-category of miracle-working images in late medieval and early modern Europe.11 Like other sacred images, the image’s sacred power was endowed neither by the material substrate of the icon nor the pictorial form of the image—tantamount to idolatry—but rather was bestowed by the divine prototype. The icon was only ever a copy of a divine original, through whose ‘cooperation’ the image became infused with the subject’s supernatural, miracle-working powers.

What was particular to images attributed to Luke was the belief that the image had been painted ‘from the life,’ from the living model of Mary and Christ. The logic of medieval and early modern theories of vision considered sight to involve the material, if invisible, imprint of the viewed object upon first the eye and subsequently the soul. Copies stood in the line of direct, material transmission of images of Mary and Christ painted by Luke’s hand. In some versions of the legend, the portraits were believed to have been touched by the living hand of their subjects, producing a contact relic. The work of eye and hand was fundamental to the process. From artist’s eye to artist’s hand, from hand to image, from image to the eye of the beholder, a viewer standing before a Saint Luke portrait was delivered the promise of the uninterrupted visual and manual transmission of the perceptual experience of an eye-witness to the holiness of Christ and Mary. While in some versions of the Luke legend the painting was completed by an angel, thus producing a likeness ‘not made by human hand’ and rendering the image itself a divine wonder, sixteenth-century sources resolutely emphasize ‘the hand of Saint Luke.’

Images attributed to Saint Luke belong to a category of miraculous images whose sacred power is undiminished through copying. In the wake of the pioneering work of Hans Belting, scholars have turned to look more closely at how miraculous images functioned as social objects and to investigate the subtle interplay of prototype and copy in the development of image cults (Belting 1994; Holmes 2013, 2018; Garnett and Rosser 2013; Casper 2021). In the words of Belting, the form of a Luke portrait icon “was not tied to a particular example but was interchangeable and repeatable” (Belting 1994, p. 47). This belief was key to the proliferation of copies of miraculous Roman icons, the Madonna of Santa Maria Maggiore included, in the twelfth and thirteenth centuries in the towns around Rome.

Recent work has explored the nuances of how copies functioned within local economies of sacred power where access to the original was frequently restricted. In the case of miraculous images, an image was not merely a convenient site from which to summon the intercession of the divine figure represented. The image itself was materially infused with the divine presence of the miracle-working holy being (Garnett and Rosser 2013, pp. 192–96; Holmes 2018, pp. 28–31). As Jane Garnett and Gervase Rosser have suggested, “the copy of a miraculous picture or statue … is understood by devotees to participate in the same qualities of divine presence and potency as the prototype image” (Garnett and Rosser 2013, p. 193).

Copies also had certain advantages over prototypes (Freedberg 1989, pp. 120–28; Garnett and Rosser 2013, p. 195). They were mobile rather than fixed, and their movement could be controlled by their possessor. They also enabled repetition, allowing the image’s power to be redeployed and stock-piled, much in the same way as prayer said with the rosary. The circulation of stories and reproductions of miracle-working images contributed further to their familiarity and notoriety, triggering mechanisms of social emulation that increased their value and desirability.

When attached to the story of Azevedo and the Brazil Martyrs, the attribution to ‘Saint Luke’ of the Madonna of Santa Maria Maggiore acquired added significance. It was the sight of Azevedo venerating the icon that incited the Calvinist corsairs to violence in the first place, imbricating the image in currents of Calvinist iconoclasm and iconophobia. The legend of Luke the artist had been weaponized centuries earlier in the iconoclasm controversy of eight-century Byzantium, where Luke attributions had been used to prove the antiquity of Christian veneration of images against iconoclast charges of idolatry, and to authenticate the verisimilitude of actual icons to the living persons of Christ and Mary (Bacci 1998; Belting 1994; Raynor 2012, 2015). Despite the many layers of anachronism—Luke was born after Christ and met the Son of God as neither infant nor adult—the legend stuck and was used to legitimate the proliferation of Luke images and copies made in their likeness. The Luke legend also struck an epistemological register. Luke’s status as an apostolic witness, the author of the Acts of the Apostles, together with his profession as a physician, augmented the purported verisimilitude of likenesses from his hand (Raynor 2015). By the early modern period, it was a truism that Luke had painted Madonna and the child Christ as living models, a theme taken up as a motif in numerous images that depicted Luke as a practicing artist (Boeckl 2005; Libina 2019). By the time Azevedo met his demise, the example of Luke the artist was wielded by Catholic theologians in defence of the veneration of images against Protestant critics.

2. Lost at Sea

The great paradox of the fortune of the Saint Luke Madonna of Santa Maria Maggiore as a globalized object is that it was the very failure of Azevedo’s copies to reach Brazil that propelled the icon into planetary orbit. The workings of sacred history and the movement of sacred objects had their own logic and followed their own momentum in the early modern world. From the beginning, the fortune of the copies of the Saint Luke Madonna aboard the Santiago were fused with the story of the massacre of the Brazil missionaries. Within months of the death of the Jesuit missionaries, an informal martyr’s cult began to solidify around the ill-fated mission. We can now turn to consider the reports of the demise of the Jesuit party that circulated in the immediate aftermath of the catastrophe. It is another irony of the story that of all the Brazil Martyrs, only Azevedo had ever set foot in Brazil.

Azevedo assembled one of the largest missionary expeditions the Society of Jesus had witnessed.12 Sailing from Lisbon with the Portuguese fleet in June 1570, the missionaries traveled on two separate ships. Some thirty-nine Jesuits joined Azevedo aboard the Santiago, while the remainder of the party traveled on a separate ship with the new Portuguese governor of Brazil. The missionaries made their way first to Madeira, where they remained for several weeks. Departing from Madeira on 30 June, the Santiago was captured by French Calvinist corsairs, led by Jacques Soria, off the coast of Las Palmas in the Canary Islands on 15 July. The corsairs killed 39 Jesuits and a relative of the ship’s captain mistaken for a Jesuit novice. Only one Jesuit survived, João Sanches, whom we will soon discuss further.

The Jesuit Pero Dias wrote to Lisbon on 20 August to report the catastrophe. Dias was with the small group of Jesuits traveling on the second ship that remained on Madeira. His description of events was based on the testimony of two Portuguese prisoners aboard the French ship who escaped to the island. By 1570, the Society of Jesus had developed an efficient and well-organized communications circuit. Once the letter of Dias arrived in Lisbon, news of the disaster spread quickly through the Portuguese colleges. On 18 September, the Lisbon procurator forwarded the letter to Rome, where it was translated into Italian in the Jesuit curia. According to the registers kept by secretary Polanco and his assistants, on 13 December copies of the account of “the 40 Martyrs of India” were dispatched to Jesuit colleges in Italy and northern Europe. Together with letters from the Jesuit Asian missions, the account of Dias was sent to a trusted Roman printer, Antonio Blado. It appeared in Blado’s Nuovi avisi that same year.13

A longer narrative of the massacre was written in 1571, derived from the testimony of João Sanches, the sole Jesuit survivor of the massacre. Sanches was a Jesuit lay brother spared by the corsairs seemingly due to his abilities as a cook. He remained on the French ship until it returned to La Rochelle. From there, he made his way to the Jesuit college at Ocaña in northeast Spain, and subsequently to Portugal. In Lisbon his testimony was recorded (and embellished) by Gaspar Mauriçio Serpe, rector of the college of Santo Antão. Serpe’s “Enformação” or “História” offers a summary account of Azevedo’s journey from Rome in 1569, a report on the voyage from Porto to Madeira, and a lengthy account of the massacre aboard the Santiago. Sometime after 1574, Serpe composed several complementary chapters (missing in some manuscript copies) describing the activities of Azevedo and the mission party during their lengthy sojourn in Portugal prior to their departure.14 Though inflected with overtones of the martyr’s cult already developing around Azevedo and his companions, Serpe was on the ground in Lisbon in 1569–70 and was closely involved with preparations for the mission.

Before turning to examine the description of events aboard the Santiago in these sources, we can first consider what Dias and Serpe have to say about copies of the Saint Luke Madonna. Their testimony includes reference to the copies produced in Portugal. Dias mentions two copies. The first is the copy Azevedo held as he lay dying: “an Image, which he held in his hands, which was a likeness of the Image of Our Lady in Santa Maria Maggiore made by Saint Luke, that he brought from Rome in a painting on copper (huã lamine de cobre), to which he was most devout”.15 The second copy was a painting on canvas: the French corsairs “also took from Father [Azevedo] an Image of Our Lady on canvas taken from the original (tirado pollo natural) of Santa Maria Maggiore in Rome, made by brother João de Mayorga, who came with him from Aragon”.16

The letter of Dias is the only source to state that the image held by Azevedo came from Rome. It is possible that the testimony of Dias on the Roman provenance of the icon was coloured by the presence of other Roman objects—rosary beads, Agnus Dei, relics, and so on—that accompanied the missionaries. Before joining the mission, Dias had served as procurator of the Jesuit college at Coimbra and would have been aware that Lisbon was well-stocked with Roman devotional objects, images included, to supply the overseas missions.17 Moreover, Dias’s testimony on the second copy is fraught with apparent contradiction. He asserts both that it was copied from the Roman original at Santa Maria Maggiore and that it was the work of João de Mayorga, who had joined the missionary expedition at Valencia in 1569 during Azevedo’s return journey from Rome. The incongruencies are perhaps resolved in light of early modern notions of the nature of Saint Luke portraits, in which distinctions between copy and prototype were unimportant to the spiritual power of the image. In this sense, ‘Rome’ may have functioned as a spiritual attribute of prototypes and copies alike, wherever they were made.

For enlightenment, we can turn to Serpe. The 1571 “Enformação” mentions the Saint Luke Madonna several times. Like Dias (whose text Serpe presumably followed in this detail), Sanches told Serpe that the image Azevedo held in his hands when murdered was a painting on copper. Sanches stated (perhaps under questioning from Serpe, cross-checking the testimony of Dias) that another painting of the image was kept by Jacques Soria due to its value. This perhaps corresponds to the copy on canvas mentioned by Dias. Sanches also testified, somewhat cryptically, that Soria kept an additional copy in marble of the “face” of the St. Luke Madonna. In a lengthy account of the shipboard drama spiked with extraneous detail, nowhere does Serpe state that any of the likenesses aboard the Santiago came from Rome.18

The preliminary section of Serpe’s “Enformação” provides additional information on Azevedo’s journey from Rome in 1569 and the production of copies of the icon. Serpe states that among the gifts given to Azevedo by the pope were a plenary indulgence for all who joined the Brazil mission and several relics. Among the relics was the head of one of the Eleven Thousand Virgins of Cologne, a significant detail to which we shall return. He also mentions “an altarpiece (retavolo) of the image of Our Lady painted by the venerable Saint Luke,” the gift of Borja for Catherine of Portugal. No other Roman copy is mentioned.19 Crucially, Serpe records several copies made by Mayorga in Portugal:

Among those he [Azevedo] brought with him was an excellent painter, a brother from the Province of Aragon. While he was here, he made four copies of the image of Our Lady, all very well done: one for Brazil, another for the college at Coimbra, another for the college of Évora, and another for that of Santo Antão.20

Serpe also states that Mayorga, together with a Jesuit carpenter, Antonio Ferz, and another lay brother, Alonso de Bayena, undertook work for the Jesuit college at Funchal on Madeira, where the trio produced several devotional objects and “an image of Our Lady taken from that of Saint Luke”.21 In light of the conflicting testimony of Dias and Serpe concerning the copy or copies of the Saint Luke Madonna made by Mayorga, it is possible that the painting on copper was similarly copied from the ‘original’ Azevedo actually did bring from Rome, namely the painting presented to Catherine of Portugal. The uncertain nature of the evidence is enough to cast doubt on the Roman provenance of any of the copies on the Santiago. Their status as gifts from either Borja or Pius V becomes even more questionable.

However many copies of the Saint Luke Madonna were aboard the Santiago, and wherever they originated, the icon was central to the narrative of Azevedo’s martyrdom. Dias described how Azevedo consoled terrified passengers with the image. He then related the story of Azevedo’s death:

The first they killed was Father Ináçio Azevedo. Seeing him with the image in his hand, they said angrily that they [the French corsairs] were Catholics, and they [Azevedo and his companions] were Lutherans and heretics, and other things, and dealt him a blow to the head, which soaked the image he held in his hands in blood, which was a likeness of the image of the Madonna in Santa Maria Maggiore made by Saint Luke, which he had brought from Rome in a painting on copper, to which he was much devoted. Afterward they dealt him two [more] blows. Wanting to take the image he held in his hands, they were unable to do so. While embracing Father Diogo de Andrade, they killed both of them and threw them into the sea with the image in his hands.22

Composed barely one month after the massacre occurred, the account of Dias is ripe with confessional tension. Devotional objects and images are foregrounded throughout. Dias recounted a mock-hanging of the skull relic of one of the Eleven Thousand Virgins of Cologne and described how rosaries, relics, books, and other devotional objects were thrown into the sea. He also recounted how the copy of the Saint Luke Madonna produced by Mayorga together with other religious images were gathered in a corner of the ship, used by the corsairs for target practice with their daggers (and introducing yet another point of divergence from the testimony of Sanches).23 The Italian translation printed in the 1570 Nuovi Avisi included all these same points. While some details were omitted, the account in the Nuovi Avisi placed even greater inflection on the fact that the Saint Luke Madonna never left Azevedo’s grasp, even as he was thrown into the sea.24 Critically, mention of João de Mayorga is elided entirely, effectively writing Portuguese copies out of the mainstream narrative.

Serpe similarly foregrounded the intersection of devotional objects and confessional violence. He described how Jesuits distributed Agnus Dei and rosaries aboard ship, providing instruction in their use. At sea, Jesuits habitually performed mass at an altar adorned with “an image of Our Lady by Saint Luke”.25 Once the ship was boarded by the French corsairs, Azevedo led Jesuit lay brothers in recitation of the litanies of the saints while holding an image of the Saint Luke Madonna, “to which he was very devout”. The Jesuits attested that they were prepared to defend the Catholic faith and the Roman Church, “and to die as good Christians … protesting aloud their Catholic faith”. While the corsairs terrorized passengers and crew, Azevedo unceasingly provided “a clear and manifest example of his faith, remaining fixed in the same place,” holding the image of the Madonna aloft and wielding the image “as an emblem of his faith, having no other arms to present to the heretics”. Even as he sustained blows, Azevedo continued to grasp the image, which became saturated with his blood, as did the Jesuit brothers embracing him, “all crying rivers of tears”. As Azevedo lay dying, his eyes remained fixed on the Virgin (Brazão 1943, pp. 553–54).

Serpe layered detail upon detail in his tale of Azevedo’s demise. As we have seen, Dias reported that the corsairs had been unable to remove the Saint Luke Madonna from Azevedo’s hands, even as he was jettisoned into the sea, accepted as one of the first miracles worked by the Jesuit. Sanches offered additional testimony that Azevedo’s arms extended in the shape of the cross as he expired. This fact was similarly considered a miracle by the Jesuit brothers who watched and venerated his body. Furthermore, when thrown into the sea, his body sank immediately, “as though it was a sack of earth, or a stone statue”. Not once, Sanches testified, did the lay brothers who witnessed Azevedo’s death remove their eyes from his corpse, right up to the moment it vanished beneath the sea. It was evidently the will of God, Serpe intoned, to allow Azevedo’s body, which in life had been a true friend of the cross, to assume the form of the cross in death (Brazão 1943, p. 559). Serpe supplies what is required of any good martyr’s account: the presence of witnesses (unfortunately martyred alongside Azevedo, but present all the same), a public declaration of faith, equanimity in the face of death, and, in the climax to the narrative, baptism by blood, tears, and finally the sea itself.

Like Dias, Serpe described the desecration of sacred objects by the Calvinist seamen. Enraged by Jesuits praying before images, the corsairs killed them upon sight. The heretics inflicted “diabolical damage” on devotional objects, throwing relics, images, rosaries, holy oil, indulgences, and papal documents into the sea. Discovering a chest of vestments and sacramental objects, they performed a mock mass using a large Agnus Dei for the host, which was then thrown to the ground and struck with daggers. Afterwards, everything was thrown into the sea.

The tale was told and retold. The letter of Dias was reprinted several times. The account was taken up in popular vernacular histories such as the Chronologia universale of Girolamo Bardi and Paul Piguerre’s Histoire de France (Bardi 1581; Piguerre 1582). It received an entry in the widely reprinted Theatre of Cruelties of the Heretics of Our Time, complete with a woodcut engraving of the murdered Azevedo afloat in the sea with a statue of the Virgin tucked under his left arm, offering yet another interpolation of the central narrative (Figure 3).

3. Martyrs and Madonnas

The devotional cult that consolidated around the doomed missionaries did not rise from the print-shop floor. The veneration of the Brazil Martyrs was, above all, an internal Jesuit cult. Many of the murdered missionaries were lay brothers (‘temporal coadjutors’ in Jesuit parlance) or young Jesuits. For this reason, the cult was particularly popular among Jesuit lay brothers and novices (Osswald and Palomo 2009, p. 132). The testimony of Dias is present in most of the surviving letter books of Jesuit colleges from this period. From Rome, it was ordered that the letter be read aloud in the colleges every July 15 to commemorate the massacre. Extant manuscripts of Serpe’s longer narrative indicate that copies were made for most of the Jesuit colleges in Portugal. Several copies bear traces of having been read aloud, likely in refectory during mealtimes. The master of novices at Évora and Coimbra each had their own copies. Serpe’s text also circulated in the Jesuit missions in Portuguese contact zones overseas (Maurício 1978, pp. 94–97).

As Serafim Leite observed many years ago, the fate of the Brazil missionaries was regarded as an act of martyrdom from the very beginning (Leite 1938, p. 264). Dias and Serpe refer to the slain Jesuits as martyrs as a matter of fact. A clause commemorating the Society’s many ‘martyrs’ in the ‘East and West Indies’ was inserted into a June 1571 bull of Pius V that recognized the Society of Jesus as a mendicant order. Though the bull had little to do with Jesuit missions, the clause was widely interpreted to refer to Azevedo and the Brazil missionaries.26 Already in 1571, among Jesuits in Portugal petitioning to be sent overseas (in the so-called Indiapetae letters in which Jesuits attested to their missionary vocation), the example of Azevedo and the Brazil Martyrs was not infrequently summoned (Osswald and Palomo 2009, p. 133).

It was only after Azevedo’s death that copies of the Saint Luke Madonna began to flow in earnest. While Borja sent several copies to members of European dynastic élites, this was a closed circuit of circulation limited to figures with whom Borja, who had moved in Spanish court circles before becoming a Jesuit, enjoyed personal contact. Others were connected by blood or marriage to Philip II.27 So, too, the first copies sent to Jesuit establishments were devotional gifts sent by Dionisio Vázquez, Borja’s biographer whom we met above. Vázquez enjoyed a close spiritual relationship with Borja and served as his confessor in Rome. The copies, authorized by Borja as a spiritual favour to Vázquez, were dispatched in December 1569 to Jesuits in Seville and Peru with whom Vázquez had personal ties. This, too, was a closed circuit.28 The copies all left Rome before Azevedo’s death on 15 July 1570. There is no sign of a concerted campaign by either Borja or Pius V, as is ceaselessly suggested, to launch the Madonna of Santa Maria Maggiore into global orbit.

Demand for the Roman icon, and in many cases the supply of copies as well, was fueled by Jesuits outside Rome in conjunction with the rise of the cult of the Brazil Martyrs. Two episodes offer precious clues. The first bears on the first documented copies of the image to appear in Jesuit establishments in Spain subsequent to the copy sent by Vázquez to Seville in 1569. A copy of the icon was initially obtained by the Spanish Jesuit Baltasar Álvarez, dispatched to Rome in 1571 to attend the tri-annual meeting of Jesuit procurators. Álvarez, Master of Novices at Medina del Campo, had already obtained a reputation for his spiritual fervour and mystical vocation. He visited the Holy House at Loreto and other sacred sites during his Italian sojourn, and departed Rome with a copy of the Saint Luke Madonna in hand. How it was obtained is not known. In Spain, Álvarez installed the image in the refectory of the Jesuit house for novices at Medina del Campo. Soon afterwards, copies of the Medina del Campo image were made by the Jesuit artist and architect Giuseppe Valeriano. Though Italian, Valeriano joined the Society in 1574 while employed as an artist in Spain. By 1575 he was residing at the novitiate at Medina del Campo. There he made copies of the image for Jesuit colleges at Seville, Granada, and Marchena. By 1576 the ‘Roman’ copy at Medina del Campo had made its way to the new Jesuit novitiate at Villagarcía de Campos, where Álvarez served as the institution’s first rector and Master of Novices and where Valeriano was perhaps involved in the completion of the site. There the icon was placed in the chapel for novices, dedicated to none other than the Virgin of Saint Luke.29 It is notable that the initial diffusion of the Madonna of Santa Maria Maggiore in Spain was achieved through locally-produced copies. The trajectory strongly suggests that early demand and reception of the icon was sustained by the spiritual appetites of Jesuit novices. In 1571, novices in Prague also received a copy of the Madonna from Rome (D’Elia 1954, pp. 304, 310).

The second episode concerns the copy of the icon that arrived in Salvador da Bahia in Brazil in 1575. One year earlier, the anniversary of the Brazil Martyrs had been observed as a feast day in the Jesuit college in that city. Epigrams were read aloud, and the example of the martyrs served as the subject of a sermon to lay brothers. In the Jesuit account of events, Azevedo and the Brazil Martyrs were celebrated as the ‘patrons’ of Brazil.30 It was in May of the following year that a copy of the Saint Luke Madonna arrived in the city, together with relics of the Eleven Thousand Virgins of Cologne. Madonna and relics found their permanent home in the Jesuit church in the city. The translatio occurred on the feast day of Corpus Christi in 1575. Icon and relics were carried in procession through the town amidst music, singing, and fireworks.31 The consignment of Cologne relics was not a casual choice. As we recall, a skull relic of one of the Cologne Virgins had traveled with Azevedo aboard the Santiago. Over the next decade, Jesuit churches in Espírito Santo and Rio de Janeiro in Brazil, and Michoacán in New Spain, similarly received copies of the Saint Luke icon in tandem with relics linked to the cult of the Eleven Thousand Virgins.32

Jesuit novices and Cologne Virgins – such are the slender threads that tie the cult of the Brazil Martyrs to the global circulation of the Madonna of Santa Maria Maggiore. The tenuous nature of the evidence penetrates to the heart of the matter: traces of the martyrs were nowhere to be found. Fortunately, there is another way to confront the problem.

4. Martyrs without Bones

The indefatigable presence of the Saint Luke Madonna aboard the Santiago was core to the unfolding of the martyr’s cult that emerged around Azevedo and the Brazil missionaries. A remarkable body of sources illuminates the early contours of the cult. Between 1628 and 1632, the Society of Jesus collected testimony on Azevedo in Portugal and Brazil.33 The hearings were in response to restrictions placed on the celebration and veneration of non-canonical saints and martyrs, introduced in a series of Holy Office decrees and papal legislation between 1625 and 1634.34 Until then, Jesuits may have been relatively sanguine about the possibility of receiving official recognition of Azevedo as a holy martyr. Francis Xavier, the first and most famous Jesuit missionary, had been canonized in 1622 together with Ignatius of Loyola, the founder of the Society of Jesus. Rome now sought to assert greater control over the canonization process, regulating new cults and creating a more complex bureaucratic apparatus. The new regulations forbid informal expressions of devotion towards non-canonical saints or martyrs, banned the veneration of their likenesses, and imposed new standards of evidence.

Sanctification was hardly a straightforward process at the best of times. In the case of Azevedo and the Brazil Martyrs, however, there was a major structural impediment to official recognition of martyrdom: the absence of bodily remains (San Juan 2008, pp. 250, 254; Osswald and Palomo 2009, pp. 137–39). Christian martyrs’ cults were intertwined in complex ways with both the rise of the cult of the saints and the cult of relics. Since late antiquity, martyrs’ tombs and martyrs’ relics had provided an important locus for the veneration of saints (Brown 1981; Wiśniewski 2018). Bodily remains provided both physical evidence of holy sacrifice and a material locus for engagement with a martyr’s divine power. Under the new canonization rules, the prospects for Azevedo’s cause looked bleak. The lack of physical evidence would plague proceedings for centuries. The impossible spectre of a martyr’s cult without martyr’s bones hung over the entire undertaking.

The absence of relics had been a devotional predicament before it became a juridical conundrum. For those who did not doubt the fact of the martyrdom of the Brazil missionaries, the issue was not the absence of material proof but the lack of requisite physical materials and topographic sites—visible, tactile, visitable—necessary for the holy martyrs to manifest their divine presence. There were a handful of contact relics, such as the chalice with which Azevedo was said to have celebrated his last mass on Madeira and that purportedly bears his teeth marks. However, they were few in number, largely unknown, and inaccessible to all but locals (Osswald and Palomo 2009, pp. 137–39). Such was the paucity of physical matter through which the martyrs might manifest their sacred power that in later stages of the proceedings, mystical visions of Azevedo’s martyrdom were offered as evidence of the workings of the holy martyrs (San Juan 2008, pp. 253–54).

The proceedings of 1628–32 reveal a complex web of associations and proxies that functioned to supplement the missing remains of the dead Jesuits. The Saint Luke Madonna was at the centre of this devotional nexus. Nor was this the first time the icon had been summoned to supplement absent relics. As Hans Belting has suggested, one of the reasons for the genesis of the legend of Luke the artist in late antiquity was to assuage the hunger for material remains of Christ and Mary left unsatiated by their empty tombs. Icons commonly functioned as proxies for missing or distant relics (Belting 1994, pp. 59–63; Raynor 2012, pp. 68–73). Late antiquity also witnessed the coalescence of theologies of the Ascension of Christ and the Assumption of the Virgin, which posited the material transport of the physical bodies of Christ and Mary to heaven. This development was similarly intertwined with the material and social logics of the cult of relics.

The task of the proceedings was to turn devotional practice into legal proof. The new rules placed Jesuits in a difficult position. To receive acknowledgement of the legitimacy of the Azevedo cult, Jesuits were at once required to provide evidence of Azevedo’s reputation for sanctity, or his fama sanctitatis, while at the same time show that he enjoyed no public cult, that his veneration was restricted to Jesuits (Ditchfield 2007, pp. 213–14). The testimony of the proceedings is wandering, repetitive, and anachronic. This, of course, was the point: to trace the veneration of the Jesuit martyrs to the first months and years following the massacre, and to display the unwavering uniformity of sacred ‘knowledge’ and practice of the cult from one location to the next, from the hills of northern Portugal to coastal Brazil. Several older Jesuits provided testimony. Some, who would have been teen novices at the time, claimed to have met Azevedo prior to his departure. Others attested to their lived experience of the veneration of the Brazil Martyrs. The Saint Luke Madonna surfaces time and again in the proceedings.

One after another, witnesses offered detail upon detail of the massacre aboard the Santiago. The general backdrop is one of confessional violence, iconoclasm, and the persecution of true Catholics by fanatical heretics. Four recurring themes in the testimony are of significance for understanding the intersection of the Saint Luke Madonna with the cult of the Brazil Martyrs.

The first theme concerns the copy of the icon Azevedo carried aboard the Santiago. The most striking aspect of the testimony is the universal agreement of all witnesses, well before the tale was taken up by Jesuit chroniclers, that Azevedo had received the copy destined for Brazil directly from Pius V in Rome. In Braga, Coimbra, Évora, and Lisbon in Portugal, and in Salvador da Bahia in Brazil, witnesses reported as a matter of fact that Pius V gave a copy of the image to Azevedo in Rome. They stressed the rarity and exceptionality of the copy, asserting that never before, in the words of one witness from Braga, “had the pope or his predecessors permitted copies to be made of the aforementioned image of the Madonna”. Almost every reference to the image mentions Luke: “painted by Saint Luke,” is the constant refrain.35 Finally, the crucial fact from the first reports of 1570–71 is offered up repeatedly: Azevedo died with the image of the Saint Luke Madonna lodged in his hands.36 The chain of manual transmission of the image traverses time and space: from the hand of Luke the artist, to Pius V in Rome, to the hands of Azevedo. The story supplies papal authentication of the copy of the Madonna carried by Azevedo aboard the Santiago and, by association, papal validation of the martyr’s cult that subsequently arose around the image’s courier.

The second theme concerns the copies of the icon made by the Jesuit painter João de Mayorga for the Jesuit colleges in Portugal.37 Mayorga had been killed alongside Azevedo on the Santiago and was regarded as a martyr in his own right. Some witnesses asserted that the copies had been made from the copy of the image aboard the Santiago, a necessary corollary of the new fact that this was the copy Azevedo had received from Pius V in Rome. By this means, copies of the icon in Portuguese colleges were transformed into mute witnesses to the existence of the image destroyed by Calvinist iconoclasts, which now gained the status of a lost, secondary prototype. In the case of the copy made for the college of Santo Antão in Lisbon, it was stated that Azevedo himself had delivered the image to the college, producing a contact relic. Here is the evidence of one witness from Évora:

He said that he knew that the aforementioned Father Ignazio d’Azevedo was received with great benevolence by the said Pope Pius V, from whom he obtained license to have a copy made of the Madonna of Saint Luke, which out of piety had never been copied, and Father Ignazio d’Azevedo gave a copy of the said Image to the College of Santo Antão, which is venerated as a relic, having passed via the hand of the said father … [where] it is seen today in the brothers’ chapel of the aforementioned college.38

Amidst yet another variation of the origin story, the mention of the hand of Azevedo is hardly casual.

Third, the Saint Luke Madonna is frequently paired with another object Azevedo actually did carry from Rome: the skull relic of one of the Eleven Thousand Virgins of Cologne.39 One older Jesuit testified that before his departure, Azevedo had visited Évora, where he displayed the image of the Holy Virgin together with the head of one of the Virgin Martyrs to members of the college.40 We have already seen that the early transit of copies of the image to Jesuit colleges was at times paired with relics of the Cologne martyrs. As Rose Marie San Juan has suggested, Jesuit appropriation of the cult of the Eleven Thousand Virgins was connected to the nascent cult of the Brazil Martyrs (San Juan 2017, pp. 418–22). There were important congruencies between the two martyrs’ cults. Though the Cologne Virgins did not perish at sea, there were time-honoured maritime associations of Saint Ursula and the Virgins, who, according to legend, sailed from Roman Britain to Cologne. Ursula converted many of her pagan companions, and undertook a pilgrimage to Rome from Cologne before enduring martyrdom upon returning to the German city. At Cologne the Virgins endured slaughter at the hands of Vandal marauders, who were incensed by their Christian faith. Like the Cologne Virgins, Azevedo was martyred by non-believers after a journey to Rome; had endured the hardship of maritime travel; and, together with thirty-nine others, had experienced martyrdom as a corporate body. The Eleven Thousand Virgins first and foremost provided a canonically authorized example of collective martyrdom. Moreover, the incorporation of relics of the Eleven Thousand Virgins into Jesuit devotional spaces provided a material locus for the veneration of the Brazil Martyrs.

Finally, witnesses repeatedly testified to the long-standing veneration of images of Azevedo and the Brazil Martyrs. The testimony was tilting at the recent strictures placed on the display and veneration of images of non-canonical saints and martyrs. Again and again, witnesses testified to the long-established practice of venerating images of the martyrs at “private altars” on the anniversary of their death.41 We can consider the testimony of one older Jesuit from Salvador da Bahia, Brazil:

He said that in the forty-three years that this witness was in the Society of Jesus after the martyrdom of these servants of God, they were always held in honour and repute and venerated as Martyrs of Christ, and that their printed and painted images were displayed in the chapels of the college here in Bahia, and in the chapel of the novitiate in Coimbra, by way of ornamentation of the said chapels.42

As we recall, a copy of the Saint Luke Madonna had arrived in Salvador da Bahia in 1575, together with relics of the Eleven Thousand Virgins of Cologne. By 1583 the Madonna had been installed at one altar, three skull relics of the Cologne Virgins at another, and a fragment of the true cross at a third (Cardim 1925, pp. 288–89). According to this same witness, the images of Azevedo and the Brazil Martyrs were normally kept in the sacristy at Bahia and were only brought out to the chapel on their feast day (15 July) and on other holy days. In other words, the portraits of Azevedo and the Brazil Martyrs were paraded in procession from the sacristy to ‘meet’ the Saint Luke Madonna and relics of the Eleven Thousand Virgins, their companions aboard the Santiago. The witness attested that this was common practice across Jesuit establishments in Brazil and in residences for novices in Portugal.43 What was more, it all happened behind closed doors, without a whiff of illicit public spectacle.

Such was the strength of the association of the Saint Luke Madonna with Azevedo that it is hardly surprising that the copy of the icon at Salvador da Bahia was believed to be the same blood-soaked copy that Azevedo had held in his hands as he perished. In one version of the legend that ingeniously combines the accounts of Dias and Serpe, Azevedo’s body, in the shape of the cross and clutching the image of the Virgin, sunk to the bottom of the sea where it remained for an indeterminate period of time. The holy corpse rushed to the surface when it sensed a Catholic ship passing above. At the moment he opened his mouth to utter the name of Jesus—for the third time since his death—Azevedo opened his hand to release the sacred image imprinted with the martyr’s blood-stained fingerprints. It was subsequently carried to Salvador da Bahia, where it was venerated “enameled in blood, and bathed in glory”. By means of this miracle, the Holy Virgin completed her voyage as Azevedo had not.44

Azevedo’s devotion to the Saint Luke Madonna provided implicit justification for the veneration of images of the Brazil Martyrs. This logic finds iconographical confirmation in the frequent inclusion of likenesses of the Saint Luke Madonna in portraits of Azevedo and the Brazil missionaries.45 The conceit of including an ancient icon within a contemporary group portrait authorized viewers to transfer Azevedo’s veneration of the Madonna to the primary subject of the painting, the Brazil Martyrs. The example seen here is from the Catedral de Santa Ana at Las Palmas de Gran Canaria, near where the massacre occurred and where the missionaries were celebrated as the ‘martyrs of Tazacorte’ (see Figure 4).46 In strictly geographical terms, the painting was as close to a martyr’s tomb as could be hoped for.

Images and relics worked in tandem to provide a material nexus for interaction with the holy martyrs. There is constant slippage in terminology in the sources, a recurring transfer of sacrality from one object and place to another. The notion of the Saint Luke Madonna as an Azevedo contact relic was likely not restricted to the Portuguese copies of the icon made by João de Mayorga, but was extended to the ‘faithful’ likenesses venerated in Jesuit chapels and churches across the globe. Sight, too, was tactile: the image was the very same as that both held and beheld by the dying Azevedo. In a similar fashion, the inexhaustible supply of relics of the Eleven Thousand Virgins that populated these same sacred spaces had once been interred alongside the virgin’s skull martyred a second time aboard the Santiago. Relics of the Cologne Virgins, copies of the Saint Luke Madonna, and portraits of Azevedo and the Brazil Martyrs operated as proxies for the missing bones of the martyred Jesuits, each augmenting the power of the other. On one level, the Saint Luke Madonna affirmed the dubious verisimilitude of painted likenesses of the Brazil Martyrs. More powerfully, copies of the icon provided a tangible link to the martyrs’ incarnate presence on earth through an image localized in geographic space not only in Rome, but in Lisbon, Coimbra, and Évora in Portugal, in Salvador da Bahia and Rio da Janeiro in Brazil, in Lima in Peru, in Michoacán, Patzcuaro, and Guadalajara in New Spain, at Goa and Fatehpur Sikri in India, in Arie in Japan, and at Gorgora and Aksum in Ethiopia.

And so, the Madonna of Santa Maria Maggiore became not only a globally dispersed image but an emblem of global Catholicism. At the same time, the image moved within a narrowly circumscribed circuit. The image was laden with associations with mission, martyrdom, and the defence and propagation of the faith. Neither Francisco Borja nor Pius V propelled the image into global transit. It was the everyday devotional logic of a Jesuit martyrs’ cult that pulled the image from Rome.