The Society of Jesus in Peru was summoned by Philip II to impose Tridentine Catholicism, independently of Rome, a task that entailed compliance with the colonial authorities’ demands. In 1568 Philip II imposed a reorganization of all the aspects of the mission: political, economic, religious and ecclesiastical, among others. Sent by the king to implement all these reforms, Viceroy Toledo (1569-1581) played a key role in this new stage in the colonization of Peru after the conquest. After arriving in Peru in 1569, together with the Society of Jesus and members of the Inquisition, the vast colonial enterprise was now in Toledo’s hands. Philip II had entrusted the Society of Jesus with the task of redirecting evangelization in the Andes, as he was not satisfied with the Dominicans, Franciscans, Mercedarians and Augustines who had preceded the Jesuits.

As the theologian of the mission, Acosta played a key role in adapting the Catholic doctrine to Andean soil. Before Acosta’s arrival, Philip II had imposed the decrees of the Council of Trent (1545-1563) as the unquestionable law for the colonies, which meant the imposition of Counter-Reformation norms for standardized and universal catechesis. From the first years after the conquest, around the 1540s, the Spaniards had begun to write down statistical information—e.g., census data, tribute lists—that the administrators of the vanquished Inca Empire had registered with their quipus in pre-Hispanic times. Current interpretation distinguishes between two kinds of quipus in the Inca Empire: the accounting quipu and the historical, memory or rhetorical quipu. The accounting quipu registers only numbers, but these quipus were nonetheless endowed with signs that appeared in the form of subsidiary strings attached to either the primary cord or the pendant strings that modified aspects of the items and numbers registered. The accounting quipu was used to fulfil accounting functions from Cuzco down to the smallest groups in a lordship. The historical quipu also had “signs” in the strings (subsidiary strings attached to either the primary cord or the pendant strings), but the European sources emphasized the “technical memory” of the team trained to reproduce a reading of these historical quipus. Unfortunately, many of these quipus were lost with the European invasion, when the Spaniards killed the quipucamayocs and burned their quipus. The new colonial dominion resulted in two contradictory perspectives and responses regarding the quipus. On the one hand, it needed to retain the accounting for a certain period of time in order to colonize the newly subjugated people. On the other hand, the colonial order was obliged to suppress the dangerous historical-ideological “official records”—memory quipus (Assadourian 2002: 122-34).

The general inspections (visitas) ordered in 1549 by Pedro de La Gasca, the president of the Real Audiencia, the court of appeals in Lima, precipitated an episode of systematic transcription of statistical data from quipus to written Spanish documents. The Toledan inspection procedures called for a more systematic collection of historical information than had been carried out up to the 1570s, and the documents produced as a result of these new rounds of inspection visits grew in profusion. Furthermore, they reflect the key role played by local quipucamayocs in supplying Spanish officials with statistical information, on the basis of which the newly reorganized colonial state began to take shape. However, the Toledan investigations represent the heightened tensions and conflicts between the quipus and the written records. In short, the interaction between Andean and Spanish records and record keepers was by no means always amicable, especially when the central issue became who—natives or Spaniards—would be in control of public records in the colony. Since the quipu system was too distant from European modes of expression, the Toledan reforms of the 1570s had aimed at replacing quipus with alphabetic records, but they were not very successful in this respect (Durston 2007: 69).

We will now look at the role played by the quipus, whose use was allowed—and encouraged—by the Jesuits in the doctrinas. In an annual letter to General Mercurian of 1577, José de Acosta relates:

I forgot to say how in these boys [in the doctrina of Juli] who learn the doctrine I find much more skill than I thought. In just one week some of them know how to cross themselves and also the Our Father, Hail Mary and the Creed and the Salve in the language [Spanish] so they all sang it during the procession on Sunday; and many men and women and boys and girls are all day with their quipos, [1] like students repeating a lesson. (Egaña 1958: 276)

In this same annual letter, Acosta states that in the morning, when the Indians gather in the main square in front of the church, and are divided into groups of twelve and fifteen, they recite the prayers and doctrine. In effect, they are indoctrinated, but they use their quipus so that, according to Acosta, “they remember what they are taught, as we do in writing” (Egaña 1958: 280). Clearly, the quipus were related to learning and especially remembering the doctrine. Acosta in particular relates recollection to writing. The Andeans, on the contrary, recorded their lives through the tactile and visual world connected to orality, as they were not inclined to write or use images accompanying texts to express themselves (Cummins 1998: 95).

In more general terms, the missionaries in Peru made use of pre-Hispanic Andean notation. Before the Jesuits, the Mercedarian Diego de Porres recommended using the Andean quipus so that each village could have both the resolutions of the First Lima Council in 1551 and the list of the commandments. Moreover, when the Jesuits replaced the Dominicans in Chucuito, where the doctrina of Juli was located, they might have allowed the use of quipus as an inherited custom from the Dominicans, then expelled by the royal representatives (Estenssoro 2003: 217).

The Third Lima Council made an old dream come true: one of imposing an authoritative catechism embodying a unified doctrine in colonial Peru, as well as in the suffragan dioceses. Indeed, this general catechetical corpus translated from Spanish into Quechua and Aymara for all the provinces. The Greater Catechism (Catecismo mayor) and Brief Catechism (Catecismo breve), both approved by archbishop Toribio de Mogrovejo in 1583 and printed in 1584, were later complemented by the addition of the Third Catechism, by Sermons—Tercero catecismo por sermones (1585), which was made up of a series of sermons in Spanish, also translated into Quechua and Aymara. Learning of the Third Cathecism, unlike the other two catechisms, was not compulsory.

There are two issues to be stressed before moving to the second use of the quipus in the Peru mission. First, it is important to keep in mind that the Society of Jesus imposed the administration of the sacraments in Peru, reinforcing confession and incorporating communion to consolidate an evangelization grounded on positive theology. This is one of the main contrasts we find with the previous religious orders, such as the Dominicans, who were in favour of minimum doctrinal contents for the Indians, who had to embrace Christianity gradually, saving such complex dogma like the mysteries of the Trinity and the Incarnation for later. And they confined themselves to administering the sacraments of baptism, marriage and, occasionally, confirmation. Second, the Third Catechism by Sermons sets a strong anti-idolatrous tone in the Third Lima Council corpus, the Doctrina christiana para la instrucción de los indios (Duviols 1977: 175). It contains sermons that specifically denounce Indian “idolatry,” such as the eighteenth, which explicitly focuses on breaching the first of the Ten Commandments, idolatry, mainly expressed in the worship of guacas. [2] The nineteenth sermon encourages reprimanding the devil’s ministers, the hechiceros (sorcerers) and their superstitions and vain rites. It prevents Indians from celebrating the Raymi, the Festival of the Sun, for instance, or any other celebration of their ancestors, since they worshiped the devil. The twenty-third sermon is against borracheras, drunkenness, itself a mortal sin and the principal means of destroying faith and preserving superstition and idolatry: “Drunkenness brings the man closer to a beast and, even worse than a beast, he is deprived of judgement, which is the light of his soul” (Pereña 1986: 629). It is also harmful because “it kills the faith in Jesus Christ and nurtures idolatries and devilish sects of the infidels and the devil” (637).

This being said, we can now focus on those few relics of a pre-Hispanic past that were allowed to remain. In sermon twelve, on how Indians should confess their sins in order to be forgiven, it is recommended that they record their sins with quipus. The sermon addresses them by saying:

For your confession to be good, and please God, the first thing you have to do, my son, is to make a quipo, just like you make them when you are a tambo camayo, of what you give and what people owe you. So you make a quipo of what you have done against God and against others, and how many times, if many or a few. And not only is it about deeds, but also about your thoughts ... if you desired to commit sin with a woman and you looked at her with that purpose, if you want to steal a blanket or someone’s ram. (Pereña 1986: 482)

Indeed, it was regarding confession that the Third Lima Council corpus placed stress on the use of quipus, making of their manufacturing a sine qua non for a thorough confession. But only a Christian usage of the quipus was allowed, with none other that might serve their superstitions (Castillo Arroyo 1966: 91; Vargas Ugarte 1951: 358). In this regard, the catechetical corpus went in an opposite direction to that established by the Third Lima Council (1582-1583), specifically chapter thirty-seven of the third section, in which the quipus were classified as idolatrous objects and were ordered to be destroyed:

Ban all the books that narrate or intentionally teach lecherous and obscene matters, for not only faith should be considered but also those customs that become easily corrupted.... Punish all the bishops who are in possession of these books.... And, as among the Indians, ignorant of letters, there are certain signs with different strings instead of books, that they call quipos, which contain a considerable number of testimonies of ancient superstition in which they keep the secrets of their rites, ceremonies and laws, the bishops must be sure to destroy them completely, for they are pernicious instruments. (Lisi 1990: 191)

Approximately thirty years later, regarding a religious use of quipus, the chronicler Guaman Poma de Ayala also suggested that for the holy sacrament of confession the fathers make the penitent examine their soul and conscience for a week, and that they make a quipu recording their sins (Poma de Ayala 1980: 585). To conclude, even though the Third Lima Council and its catechetical corpus meant a watershed in the doctrinal, theological and disciplinary aspects of the Catholic world in Peru, the quipus proved to be effective—in the missionaries’ eyes—for the Indians to both record their sins and remember doctrine. Although the Jesuits did not directly appropriate a local memory technique, they did so indirectly by allowing and even encouraging its use for religious purposes in the doctrinas, as can be observed in Acosta’s statements. From the missionaries’ point of view and expectations, the quipus are certainly understood to be effective. As for confession, Indians did not seek confession on their own; it was mandatory, and the moment they touched the knots and strings was one of anxiety, as they had to recite their sins in front of the priest (Harrison 2002: 282).

Soon after his arrival in China, Matteo Ricci described the three “sects”—Confucianism, Buddhism and Daoism—always presenting them as three watertight categories. Ricci distinguished Confucianism from the other two, for, in his view, it was not idolatrous. The non-idolatrous gentility Ricci implicitly ascribes to the Confucians was also a response to his idea of Confucianism as a moral system that served to govern the Empire wisely but was lacking in metaphysical or supernatural foundations. This is the basis of the synthesis he constructed between Christianity and Confucianism, as they, simply put, were compatible at a moral and ethical level, and Christianity would provide Confucianism with a supernatural base. Ricci’s interpretation was the result of his study of The Four Books (Si shu) toward the end of the 1580s, which he then translated into Latin; they were all basic books in the education of scholars trained to hold imperial posts. Together with The Four Books, The Five Classics forms a compendium that was the cornerstone of Confucian education. [3] Ricci regarded Confucius, the Sage, as “another Seneca” (un altro Seneca) and The Four Books as “good moral works” (Tacchi Venturi 1911-1913, vol. 2: 117). This perception served as an impulse for the Jesuit to compose humanistic writings proclaiming wisdom from the West.

In imperial China, the oral and the written went together in the examination regime, enhancing both literacy and memory. A written tradition of compositions was rehearsed orally by candidates to recall classical texts verbatim. For those without photographic memory, instruction in mnemonic skills was part of the classical teaching repertoire in imperial China, where oral recitation was aided by rhyming characters, four character jingles, and the technique of writing antithetical pairs of characters (shu dui). Many educators in the Ming period made the memorization of these two-character phrases a major building block of a classical memory (Elman 2000: 261-63).

In a letter to General Acquaviva from Nanchang, dated November 4, 1595, Ricci says he started to teach the “memory of places” (memoria locale) to some of his visitors (Tacchi Venturi 1911-1913, vol. 2: 211). Indeed, when it became known that Matteo Ricci had a prodigious memory, and that he had mastered a Western art of memory based on remembering the order of things, the Jesuit received invitations to present his art to the literati world of the late Ming. For instance, the governor of Jiangxi province invited Ricci to teach his memory-enhancing techniques to his three sons, then preparing for the civil examinations. According to Ricci, the Chinese there soon learned that Ricci “knew an art of memory, something unprecedented for them, and being a city full of students and literati there could not be anything more welcome than this” (Tacchi Venturi 1911-1913, vol. 2: 235). But the composition of the Xiguo Jifa also responded to Ricci’s awareness of the importance and power of the written word in China, of which he made good use. Indeed, as he himself said, books could get to the places the Jesuits could not reach (Ricci 1942-1949, vol. 1: 198). Ricci shared with the Chinese literati a love of books and printing. All the Western books, material exponents of Renaissance culture, that the Jesuits had taken with them to China for their personal use became one of the most effective ways of gaining recognition from the Chinese literati (Standaert 2003: 371).

Before analyzing some aspects of Ricci’s treatise itself, let us now consider how different scholars have conceived the Xiguo Jifa so far. They tend to disagree when it comes to defining how Ricci’s Xiguo Jifa represented European culture and, more specifically, how imbued it was with humanistic culture. The translator of this treatise into German, the scholar Michael Lackner, claims that even though Ricci’s Xiguo Jifa is one of the most important documents of the adaptation of a speciality of occidental civilization to a Chinese context, it is anachronic, since it deals with a conventional medieval, Lullian, ars memorativa. Ricci probably knew Lullian ars magna, or at least some of its formal models: in his catechism, the Tianzhu Shiyi, usually translated as True Meaning of the Lord of Heaven, which he finished in 1596, he includes a diagram, an arbor porphyrii, used in the Western medieval tradition. [4] This does not mean that Ricci knew Lull’s method in particular. However that may be, according to Lackner, by Ricci’s lifetime, European humanism had already transformed the old art of memory in a most creative way (Lackner 1986: 201-219). Some years ago, in a co-authored article, Howard Goodman and Anthony Grafton explored the connection between memory, figuration, spatiality and pictorial representation, concluding that if Ricci had mastered Lullian science he could have recognized several tools also used in the Book of Changes (Goodman and Grafton 1990: 145). Another work relating to Ricci’s Xiguo Jifa is Jonathan Spence’s The Memory Palace of Matteo Ricci, which, despite its name, says little about the treatise itself, instead providing a thorough narrative of Ricci’s life (Spence 1985). [5]

The Xiguo Jifa is a six-chapter treatise. In the first chapter there is a notion of human memory as a storehouse in which men can store what they remember, ready to be used whenever they want. In sum, it is about storing knowledge, and not just about remembering; but storage requires a method (jifa). The second chapter relates how this art of memoria locale (memory places) was born, and here Ricci reproduces Cicero and the story of how Simonides created this method (Ricci 1965 [1596]: 17-18). From the third chapter on, Ricci introduces the details of his memory method through images. In the third chapter, he explains that there are different sizes of the architectural structures in which to place the images (i.e., large, medium and small, in an orderly fashion) (Ricci 1965 [1596]: 20). There are also different sizes of images, as well as other characteristics—i.e., whether they refer to abstract or concrete things, their brightness, etc.—to be considered.

The fourth chapter delves into the history of Chinese characters. Ricci refers to the theory of the formation of “six categories,” the six formation methods in people’s common everyday language described in the Origin of Chinese Characters (Shuowen jiezi) composed by Xu Shen: pictographic characters, indicative characters, associative compounds, phono-grams, phonetic loan characters and mutually explanatory characters (Ricci 1965 [1596]: 29). In introducing these categories and analyzing the forms and structures of Chinese characters, Ricci explains how different images can be made. The last two chapters provide examples of all the different possibilities for creating Chinese characters. Undoubtedly, Ricci saw much potential to use the Chinese characters as images that stand for the things to be memorized, to be placed in the storehouse, a memory palace or whatever the structure, in an orderly fashion. Indeed, what Ricci discovered in China, and added to this European art of memory, is that the Chinese characters themselves were, in his view, ideal images for a memory palace.

Ricci seems to ignore local mnemonic techniques and how they worked, or at least he does not mention any of them. The use of diagrams (tu) for mnemonic purposes was a rather common phenomenon in China. During the Song Dynasty (960-1279 AD), and especially in the Neo-Confucian tradition, diagrams could be a visual support for textual analysis. In short, these diagrams were based on a text or a corpus of texts. The diagram was meant to give an analytical overview of the structure of the text and was also a visual aid for memorizing it (Lackner 1996: 210-12). In this regard, the popularization of woodblock printing had opened up new opportunities for writers of different subjects like cosmology, mathematics, medicine and architecture, among others, and thus developed richer forms of communication through the use of diagrams, charts and illustrations (Bray 2007: 522-23). However, as Michael Lackner points out, Ricci does not include a text for which images could act as a visual aid for its memorization (Lackner 1996: 211). And here we may encounter a major reason to explain the low impact of Ricci’s treatise, at least as a tool and aid for the Chinese candidates when preparing for the civil service examinations, one of the main challenges being the student’s ability to recall verbatim a classical text. Ricci focused on one of the core aspects of the Western art of memory—that is, the way it constructs its images in internal mental spaces, which are connected by a dense network of relationships to images created both by words and images, especially those produced by the figurative arts (Bolzoni 2001:182). But, from the Chinese point of view, it is likely that Ricci’s method did not offer a text that could act as an anchorage for that dense network of images, thus rendering it too abstract as a method to recall a classical text. Last but not least, Ricci’s use of the Chinese characters as ideal images for a memory palace may not have worked for the Chinese readership. For one, the Chinese character allows for a more direct access to what is meant (the signified) than is the case in alphabetic writing systems, which take a more roundabout way to the signified (Lackner 1996: 212). Therefore, one possible question to be asked here is how the Chinese characters as images could become central elements to communicate meaning and narrative to the Chinese (Rasmussen 2012: 33).

This article has focused on how memory as an art learned in the European Jesuit colleges for perfecting oratory skills was used in the China and Peru missions for different purposes. It has analyzed the Indian quipus and Ricci’s Xiguo Jifa, the former a local memory technique, the latter an “exported” ars memoriae to the China mission. In the case of Peru, the Jesuits were maybe not the first ones to realize the efficacy of quipus for remembering doctrine and recording sins before confession, but they allowed and actually encouraged their use. Even though the Third Lima Council corpus, with Acosta’s imprint, harshly attacked Indian idolatry, it was forgiving and encouraging regarding the religious use of quipus. The Jesuits—and Acosta in particular—might have been tolerant simply because, in this particular aspect, they were practical: if they wanted the Indians to not repeat doctrine in parrot fashion and to be thorough when recording all their sins, time proved that quipus were effective. And this was not just because the missionaries had to discard alphabetical writing as a tool when indoctrinating the Indians. Let us remember that in early modern Europe, the Jesuits were accustomed to preaching doctrine in rural areas where the congregation was not precisely literate. In the 16th century in Europe and later, writing was not considered the sole expression of humans’ essential capacity to create memory; the same power was attributed to painted or sculpted images and to the spoken word. Writing could enter into the equation, though only in the form of public readings, since most of the preacher’s audience was illiterate, and unmediated written texts, printed or handwritten, proved completely useless (Bouza 2004: 3-16).

Matteo Ricci in the China mission told the Chinese that the method presented in his treatise was currently and widely in use in the West, and that it was very efficient (Ricci 1965 [1596]: 16). He wanted and needed to seduce the scholar-officials in the Ming Empire, with the Emperor at their apex, for the Jesuits in China were in their hands and they would decide whether they could stay or not. As a scholar from the West, Matteo Ricci took part in enlightened debates in the academies, becoming part of a network of educated male literati who were expected to master the Confucian learning and to pass the official examinations to advance a career in the bureaucratic world. But all these efforts never diverted Ricci from his main goal of conversion of the Chinese literati, an aspect that many scholars have more than once overlooked, letting the humanist and scholar overshadow the missionary that Ricci was. Now we are more than certain that Ricci was a scholar, a humanist and, above all, a missionary. That is why one cannot help but notice that the Xiguo Jifa is probably the only one among Ricci’s works in which there is no mention of the Christian God in the Chinese version (the Lord of Heaven [Tianzhu] or the Lord on High [Shang di]). Indeed, even though the Xiguo Jifa was meant to be a treatise containing a tool for the candidates for the civil examinations, Ricci learned very well how to slip in certain religious notions, something he did right from the beginning in his first treatise in Chinese, On Friendship, which is not a doctrinal text. So we may conclude that in his Xiguo Jifa Ricci took a different tack, an indirect way of reaching the Chinese literati, by helping them to remember their “sacred” texts through a Western method. It was, however, a Western method focusing on aspects of the Chinese language that may not have been useful to the Chinese themselves.

To conclude, the goals in these distant mission spaces, Peru and China, were the same: conversion, indoctrination and confession of the “gentiles.” Both Acosta and Ricci conceived of memory as an art, a technique, local or “exported,” and a means to achieve these goals, without preconceptions of which techniques might be most efficient. This might explain why, in the colonial context of the Andes, the religious use of local memory techniques was not only not condemned as idolatrous, but actually encouraged by Acosta, when in the China mission, a paradigmatic case of Jesuit accommodation to Chinese culture, Ricci’s first impulse was to circulate a memory method with a Western core.