Ramon Llull’s Doctrina pueril, written on Mallorca during the years 1274-76, provides us with a fascinating summary of what may be called the medieval world-view. Dedicated to his son, the Doctrina pueril sets out an alternative educational programme for children, which comprises not only a general introduction to Christian life as well as to the liberal arts and philosophy, but also a description of the different religions, including Judaism and Islam. In counterdistinction to many contemporary accounts, in the Doctrina pueril Llull puts forward a decidedly historical approach to religions, both with regard to the microscopic and macroscopic structure of the work. As can be seen by Llull’s concern with the correct historical identity of Muhammad, the encounters of Christians, Jews and Muslims during the Middle Ages contributed to the perception of the historicity of religious phenomena. But, even more than this, these encounters also contributed to the perception of history as a whole, as is suggested by the second point of our article, namely the new division of history in conformity with the proper role of Judaism within it.
The Historical Identity of Muhammad Revisited
A good example for the role of historical argumentation on the microscopic level of the Doctrina pueril1 is Llull’s presentation of Islam. Thus, in chapter 71, apart from traditional polemical arguments against the founder of Islam, we also find some bio-historical information on Muhammad. Tracing the prophet’s origins, the chapter starts by saying: “Muhammad was from a city called Medina, which is ten days from Mecca, which the Saracens venerate just as much as the Christians the Saint Sepulchre in Jerusalem. Medina and Mecca and the entire region, my son, were full of people who believed in idols and adored the sun and the moon and the wild animals and the birds, and who had no knowledge of God, and they had no king and were people of little discernment and understanding. At that time, Muhammad was a merchant and was travelling to Jerusalem, and on the road, very close to Jerusalem, there was a false Christian called Micolau, a hermit, who knew a lot about the Old and the New Law. And he told Muhammad how to become the King and Ruler of the city of Medina.”2
As has been pointed out by Óscar de la Cruz, the details given on Muhammad’s life in this passage, in particular his relation to Micolau, that is the Majorquan variant of Nicolau or Nicholas, set Llull’s account in the context of a popular legend which identifies Muhammad with the first-century heretic Nicholas, the founder of the Nicholaites.3 According to this legend, which was already in circulation in the twelfth century, Muhammad was a heretical Christian deacon, who turned his back on the papal court.4
However, departing from this legend, Llull, in the passage just quoted, identifies Nicholas as having been Muhammad’s tutor. As Óscar de la Cruz and others have remarked,5 this strand of argument is, in turn, reminiscent of the legend of Muhammad as it developed in Eastern-Christian apologetics which became very popular in the Latin West. According to this legend, Muhammad had a Christian tutor who taught him the basics of religion and co-authored the Qur’an.6 Yet, it has to be noted that Llull deliberately calls Muhammad’s tutor Nicholas, instead of using his traditional name, that is Sergius or Bahira, as it appears in other contemporary texts. The Contrarietas alfolica, for instance, which Llull seems to have known, either in Arabic or in its Latin translation, clearly refers to a “Boheira, Sergius apud nos vulgare vocatur” as Muhammad’s tutor.7
Llull’s account does not fully match either of the two Muhammad-legends; rather he seems to blend together different sources, thereby creating a hybrid legend
As a result from what has just been said, it is clear that Llull’s account does not fully match either of the two Muhammad-legends; rather he seems to blend together different sources, thereby creating a hybrid legend, which uses the figure of Nicholas from one version combining it with the motif of a Christian instructor of Muhammad from the other. Now, while this is obvious, it is far less evident why Llull should have done so.
As an answer to this question, I would like to suggest the following: seemingly the identification of Muhammad and the Christian Nicholas was historically untenable, and therefore Llull decided to integrate the Nicholas-motif into the Sergius Bahira-legend. Here I can only point out some of the reasons for this argument, which are meant to show that Llull’s position is not as eccentric as it may seem at first hand, once it is considered in the broader context of the discussion on the prophet’s historical identity.
Already in the earliest appearance of the Nicholas-Muhammad legend during the twelfth century it was referred to as problematic. Thus Peter the Venerable wrote in his Summa totius haeresis Saracenorum: “Some think that [Muhammad] was that Nicholas, one of the seven first deacons, and that the sectarian doctrine of the Nicholaites, a group named after him, a doctrine which is condemned in the Apocalypse of John, is this law of the modern Saracens.”8 And after summarizing with these words the Nicholas-Muhammad-legend, Peter remarked: “Others too conjure up other individuals and, as they are indifferent with reading and unacquainted with the history of events, so also in the other cases they hold certain false opinions”.9
In spite of the harsh criticism it was subject to, the Nicholas-Muhammadlegend persisted, flourishing throughout the Middle Ages
The same account reoccurs, almost literally, in Peter’s letter to Bernard of Clairvaux.10 Both passages show that for Peter, the Nicholas-Muhammad-legend was the result of historical ignorance (rerum gestarum ignari) combined with a general lack of information.
Other authors can be quoted alongside the Abbot of Cluny, who likewise questioned the historical reliability of the Nicholas-Muhammad-legend, for instance, Mark of Toledo in the Prologue of his Latin Qur’an translation, which dates to the beginning of the thirteenth century.11
In spite of the harsh criticism it was subject to, the Nicholas-Muhammad-legend persisted, flourishing throughout the Middle Ages. As a matter of fact, the most prominent and influential version of this legend was put together in the late thirteenth century in the so-called Liber Nicolai, which has recently been edited. The anonymous author writes: “We read in the histories of the Romans that Nicholas, who was called Muhammad, was one of the seven cardinal deacons of the Church of Rome.”12 It is significant that this completely ahistorical account – as its critiques rightly point out – should start precisely with a reference to history (legimus in historiis Romanorum).
Reconsidering Llull’s account in the Doctrina pueril against the background of this contemporary debate, his identification of Nicholas and Muhammad’s tutor seems to be more than a casual contamination of two different traditions. As was the case for Peter the Venerable and Mark of Toledo, so also Llull, who was beyond any doubt much better informed with regard to Islam and its history than most of his contemporaries, could hardly subscribe to any straightforward identification of Muhammad with the Christian heretic Nicholas. At the same time, however, with texts like the Liber Nicolai, the Nicholas-Muhammad-legend had become much more established in Llull’s age than it was during the twelfth century. In this situation, converting Nicholas into Muhammad’s tutor, must have appeared to Llull an elegant way to retain elements from the wide-spread, but historically untenable Nicholas-legend.
It was probably from the Tesoro that the identification of Nicholas as Muhammad’s tutor found its way into the works of other Italian authors
At this juncture, it is worth noting that Llull was not the only author to move into this direction, even though he seems to have been one of the first, if not the very first. Thus, in other texts we observe a similar substitution of Muhammad’s alleged tutor Sergius Bahira for Nicholas, as, for example, in Brunetto Latini’s Li Livres dou Tresor.
This example is, in fact, intriguing: in Latini’s original redaction, written in France in the 1260’, the information on Muhammad is very scant. Actually, the French Tresor does not devote a chapter to Muhammad as does Llull, but limits itself to mentioning his tutor Sergius in the context of the history of the Christian Church.13 Only a few years later, however, when the Tresor was translated into Italian, its anonymous translator added substantial portions to the text, which were meant to update Latini’s account. Among these, we find a chapter on Muhammad, which starts with the following words: “A very wise and educated monk from Smyrna, who frequented the Court of Rome and whose name was Nicholas, was moved by his great faith and Christian devoutness and went to a part of Mecca, where there were only pagans who possessed no law. And he went to seek Mohammed, who was an Arab, a great man and great leader”.14 In expanding Latini’s information on Muhammad, the translator of the Tesoro integrates the figure of Nicholas into the traditional account of Muhammad’s Christian tutor, to which the French version of the work refers.
It was probably from the Tesoro that the identification of Nicholas as Muhammad’s tutor found its way into the works of other Italian authors, as for instance Jacopo da Aqui (died c. 1337). In his Imago mundi he explains that the Christian Nicholas left the Papal court and went ultra mare, where he found a socium ad male operandum, who was a diabolo ministratum, namely a merchant and camel driver named “Machometus”.15 Hence, also for Jacopo, the heretic Christian Nicholas was by no means Muhammad himself, but his tutor.
Llull decided to fuse this motif with the Sergius Bahira-legend and made Nicholas into Muhammad’s tutor, in order to arrive at a more consistent account of the prophet’s life
A third and final example of the integration of the figure of Nicholas into the Sergius Bahira-legend can be found in Fazio degli Uberti (1304-1368?) and his Dittamondo. In this text, actually not only one but several Christians are said to have served as Muhammad’s teachers:
Sergio fu l’un, del qual t’ho ditto, monico
L’altro Nicola chierico, ed apresso
Lo disperato del Papa canonico.16
This quick glance at some key texts shows that Llull’s identification of Nicholas as Muhammad’s tutor must be seen in the larger context of a historical debate about the founding figure of Islam that starts in the twelfth century, with Peter the Venerable, and continues into the Late Middle Ages and even later.17 As has been shown, the identification of the Christian heretic Nicholas with Muhammad being historically unacceptable, Llull, along with a considerable number of Italian authors, decided to fuse this motif with the Sergius Bahira-legend and made Nicholas into Muhammad’s tutor, in order to arrive at a more consistent account of the prophet’s life.
For now, and relying on the documentary evidence which is at our disposal, Llull may be considered the first author to take this line. It is noteworthy that all further evidence which we find for Nicholas as Muhammad’s tutor stems from Italian sources, all of them posterior to Llull, while there are no traces of this motif in the Hispanic and French context. It is of course hardly possible to prove a direct connection between Llull and these Italian authors; yet, one should bear in mind that Llull’s Doctrina pueril was an extremely popular text which was translated into Occitan and French shortly after its redaction and belongs to the very same genre as Brunetto Latini’s Tresor.
Leaving this question to one side and coming back to our starting point, we have to say that, on the microscopic level of his argumentation in the Doctrina pueril, the Majorcan philosopher and theologian seems indeed to have been concerned with putting forward arguments that observe certain standards of historical reasoning, thereby construing a historically more accurate account of Islam.
Towards a General Division of History: the Role of Judaism Reconsidered
The same historical approach is also very present on the macroscopic level of the Doctrina pueril, as can be gathered from the fact that the work features no less than two historical models.
The first of these occurs in chapters 68-70, under the title: “On the three laws” (De les .iii. ligs). Here Llull discusses the succession of natural law, the Old Law and the New Law. The first, i.e. natural law, covers the period ranging from Adam to Moses; the second starts with Moses, that is the reception of the Old Testament; the third, in turn, is brought about by Jesus and the New Testament. This threefold division was first outlined in the Pauline Epistles, which, modifying the Jewish tradition of Tractate Sanhedrin, assume three statuses, namely ante legem, sub lege and sub gratia.18
This brief history of paganism, Judaism and Christianity, to which one has to add the chapter on Islam discussed before, has received a considerable amount of attention in Llull-scholarship.19
The reason for Llull’s insertion of Moses, the foundational figure of Judaism, seems to result from an attempt to coordinate, his account of the three Laws with the periods of history, thereby acknowledging the role of Judaism within the history of salvation
The same is not true for the second historical model to be found in the Doctrina pueril, which occurs almost at the end of the text, in chapter 97. Under the heading “On the seven ages in which the world is divided” (De les .vii. edatz en que es departit lo mon), Llull presents and describes in fact eight (and not seven) periods of history. It has been suggested that this periodisation, which only appears in this work of Llull’s, repeats the classical Augustinian division of history into six inner-worldly periods plus two transcendent periods.20
Augustine in De civitate Dei as well as in other works, distinguished six periods or aetates of the world, drawing on the Gospel of St. Matthew (1,17): the first extends from the time of Adam to the flood, the second from the flood to Abraham, the third from Abraham to David, the fourth from David to the Babylonian exile, the fifth from the exile to Jesus and the sixth from Jesus to the end of the world; to these, Augustine added a seventh and an eight period, namely the time from the end of the world to the resurrection and, subsequently, the eternal life in God.21
Augustine’s periodisation became standard theory for many centuries; it received its most concise and influential expression in Isidore of Sevilla, who, in his Etymologiae, lists the following inner-worldly periods: “The first age goes from Adam to Noah, the second from Noah to Abraham, the third from Abraham to David, the fourth from David to Judah’s exile in Babylon, the fifth from there to the advent of the Saviour becoming flesh, the sixth, which is now, lasts until this world ends.”22
Now, looking at Llull’s account, we find that he does indeed take this description as a general frame for his division of periods; yet, to the six inner-worldly periods he adds a seventh age of the world. According to Llull, the first period goes from Adam to Noah, the second from Noah to Abraham. Until this point Llull clearly follows the Augustinian tradition as handed down by Isidore of Sevilla and others. As for the third period, which extends traditionally from Abraham to David, however, Llull splits it into two by introducing the figure of Moses. Consequently, his third period does not go from Abraham to David, but only from Abraham to Moses, while the fourth runs from Moses to David. From here on, Llull again follows the Augustinian model up to the time of Jesus.
Compared to the traditional division of history, one has to say therefore that Llull’s account actually inserts an additional inner-worldly period, which goes from Moses to David and which increases the number of periods of secular history to seven as against six, plus one transcendent period which is to come after the end of the world.23 The reason for Llull’s insertion of Moses, the foundational figure of Judaism, seems to result from an attempt to coordinate, or literally to synchronize, his above-mentioned account of the three Laws with the periods of history, thereby acknowledging the role of Judaism within the history of salvation: Llull’s first two periods, namely from Adam to Noah and from Abraham to Moses, correspond to the status ante legem or sub lege naturali, Moses, the founding father of Jewish religion, marks the beginning of the era sub lege veteri, which leads up to Jesus Christ, who, in turn, is the starting point of the era sub lege nova, i.e. Christianity.
Llull deliberately abandons the dominant Augustinian division of six world ages in an attempt to draw a coherent picture of the periods of time that should be capable of integrating his sketch of a history of religions by granting Judaism its proper place
Accordingly, it is not accurate to affirm that Llull simply repeats the Augustinian doctrine of the sex aetates mundi, as has been claimed;24 on the contrary, Llull deliberately abandons the dominant Augustinian division of six world ages in an attempt to draw a coherent picture of the periods of time that should be capable of integrating his sketch of a history of religions by granting Judaism its proper place.
In sum, our micro- and macroscopic examination of the Doctrina pueril yields the following conclusions: Doubtlessly, the encounters of Christians, Jews and Muslims during the Middle Ages contributed to the perception of the historicity of religious phenomena, as can be seen by Llull’s concern with the correct historical identity of Muhammad. But, even more than this, these encounters also contributed to the perception of history as a whole, as is suggested by our second point, namely the new division of history in conformity with the proper role of Judaism within it.25
Dialogue with, and polemics against, representatives of others faiths, such as Judaism and Islam, obliged Christian theologians and philosophers to reconsider the historical and cultural dimensions of religion. With time, the awareness of the historicity of the religions – referring to other faiths as well as to one’s own – became a decisive tool for dealing with the challenges of religious diversity. Thus, the idea of the historicity of religion, which emerges in the Middle Ages, with authors like Llull, Juan de Segovia26 and others, and which was further developed during the Age of Enlightenment, points to the need to conceive of the dialogue between the religions not simply as an encounter played out between faith and reason, but to take into account that religions are part of complex historical and social realities. Or, to put it in Ramon Llull’s words: «People [are] rooted in the faith in which they [find] themselves and in which they [have been] raised by their parents and ancestors […] And this is why, as soon as one starts discussing with them, showing them the error of their ways, they immediately scorn everything one tells them, saying that they want to live and die in the faith their parents and ancestors gave them.»27
 The original text, written in Catalan, is available in a critical edition by Joan Santanach: Ramon Llull, Doctrina pueril (NEORL VII), Palma, Patronat Ramon Llull, 2005. Its Latin version(s) have been edited recently by Jaume Medina in Raimundi Lulli Opera Latina 33 (= CCCM 215), Turnhout, Brepols, 2010.
 Ramon Llull, Doctrina pueril, op. cit., p. 182. For the Latin version(s) of this passage, see Medina’s edition in ROL 33, pp. 348-349.
 Cf. Óscar de la Cruz, “La información sobre Mahoma en la Doctrina pueril de Ramon Llull”, Taula Quaderns de pensament 37 (2002), pp. 37-49.
 Cf. the classical study by Alessandro d’Ancona, which has now been reprinted: La leggenda di Maometto in Occidente, Rome, Salerno Editrice, 1994 (1889), pp. 65-83; for an up-date of the status quaestionis, Alberto Ferreiro, “Simon Magus, Nicolas of Antioch, and Muhammad”, Church History 72/1 (2003), pp. 53-70, esp. pp. 62-70.
 In addition to the article mentioned in footnote 3, see Jordi Pardo, “Mahoma y el Anticristo en la obra de Ramon Llull”, Anales del Seminario de Historia de la Filosofía 22 (2005), pp. 159-175.
 There exists an abundant literature on this legend. The most complete and in-depth analysis is Barbara Roggema’s The Legend of Sergius Bahira. Eastern Christian Apologetics and Apocalyptic in Response to Islam, Leiden/Boston, Brill, 2009.
 Cf. Paris, BnF, ms. lat. 3394, fol. 243v; for Llull’s familiarity with this book, see Thomas E. Burman, “The Influence of the Apology of al-Kindi and Contrarietas alfolica on Ramon Lull’s Late Religious Polemics, 1305-1313”, Mediaeval Studies 53 (1991), pp. 197-228.
 English translation from James Kritzeck, Peter the Venerable on Islam, Princeton, NJ, Princeton University Press, 1964, p. 124. For the original text see Petrus Venerabilis, Schriften zum Islam, ed. Reinhold Glei, Altenberge, CIS-Verlag, 1985, p. 4.
 Cf. Patrologia Latina 189, col. 340.
 Mark of Toledo, Prologus: “Hic fuit Mahometus, id est gratiosus, non Nicolaus, falso plures autumant, cum sexcentis annis elapsis post adventum Iesu Christi hic venisset in mundum ad praedicandum arabibus, de Mecca oriundus, quae est in Arabia.” Cf. the edition of Marc of Toledo’s Qur’an translation prepared by Nàdia Petrus:Alchoranus Latinus, quem transtulit Marcus canonicus Toletanus: estudio y edición crítica, PhD-thesis, Universitat Autònoma de Barcelona, 2008, p. 7.
 Cf. the edition in Fernando González Muñoz, “Liber Nycholay. La leyenda de Mahoma y el Cardenal Nicolás”, al-Qantara 25 (2004), pp. 5-43, here p. 8.
 Brunetto Latini, Li livres dou tresor, ed. Polycarpe Chabaille, Paris, Imprimerie impériale, 1863, p. 83: “I fu li mauvais preeschierres, qui fu moines qui ot non Sergius, li quels les restraist de la foi et les mist en mauvaise error.”
 Quoted by Adolfo Mussafia in Thor Sundby, Della vita e delle opere di Brunetto Latini, Florence, Successori Le Monnier, 1884, p. 382.
 Quoted by D’Ancona, La leggenda di Maometto in Occidente, op. cit., p. 126.
 Quoted ibid., p. 82.
 As late as during the seventeenth century, we still find the German jurist Gottfried Nikolaus Ittig (1645-1710) complaining about the badly informed “glossatorem […] corporis canonici qui Nicholam Mahometum fuisse dicit” (cf. his Acta eruditorum Lipsiensium quoted by D’Ancona, La leggenda di Maometto in Occidente, op. cit., p. 122).
 Cf., in particular, the Epistle to the Romans, as well as Augustine’s Exposition of the Epistle. For the Jewish background of this doctrine, see Babylonian Talmud, Tractate Sanhedrin, f. 97a: “The disciples of Elijah taught: The world will continue for six thousand years, the first two thousand of which were a chaos (tohu), the second two thousand were of wisdom, and the third two thousand are the days of the Messiah.”
 Cf., for instance, Armand Llinarès, “Une histoire sommaire des religions selon la Doctrina pueril de Ramon Llull”, Revista Catalana de Teologia 19 (1994), pp. 99-107. See also Eusebi Colomer, El pensament als Països catalans durant l’Edat Mitjana i el Renaixement, Barcelona, Institut d’Estudis Catalans/Publicacions de l’Abadia de Montserrat, 1997, pp. 206-211.
 Cf. Josep Perarnau, “El Llibre contra Anticrist de Ramon Llull. Edició i estudi del text”, Arxiu de Textos Catalans Antics 9 (1990), pp. 7-182, here esp. pp. 35-41 (“L’Anticrist i la història”).
 See, e.g., Augustine, De civitate Dei XXII, 30.
 Isidore of Sevilla, Etymologiae V, 38.
 While the addition of Moses can also be observed in some other texts, Llull does not seem to follow them, since these usually stick to the number of six periods of secular history by not mentioning the Babylonian exile. For a general overview on the evolution of the doctrine of the six ages, see Roderich Schmidt, “Aetates mundi. Die Weltalter als Gliederungsprinzip der Geschichte”, Zeitschrift für Kirchengeschichte 67 (1955-1956), pp. 288-317.
 Cf. again Perarnau, “El Llibre contra Anticrist de Ramon Llull. Edició i estudi del text”, op. cit., p. 36.
 For further reflections, pointing into the same direction, see Óscar de la Cruz’s pages on “Man and History” in Alexander Fidora / Josep Enric Rubio (eds.), Raimundus Lullus. An Introduction to his Life, Works and Thought (ROL – Supplementum Lullianum II), Turnhout, Brepols, 2008, pp. 432-459.
 See Jesús Santiago Madrigal Terrazas, “Judíos, moros y cristianos. La visión teológica de Juan de Segovia (1393-1458) acerca de las tres culturas ibéricas”, in Matthias M. Tischler / Alexander Fidora (eds.), Christlicher Norden – Muslimischer Süden. Ansprüche und Wirklichkeiten von Christen, Juden und Muslimen auf der Iberischen Halbinsel im Hoch- und Spätmittelalter, Münster i. W., Aschendorff, 2011, especially ch. 4: “Una lectura teológica de la historia religiosa de la humanidad”.
 English translation of Llull’s Book of the Gentile and the Three Wise Men from: Selected Works of Ramon Llull (1232-1316), transl. by Anthony Bonner, Princeton, NJ, Princeton University Press, 1986, p. 302. The original Catalan text has been edited by the same: Ramon Llull, Llibre del gentil e dels tres savis (NEORL II), Palma, Patronat Ramon Llull, 1993, p. 207.