TITLE:  The Beatus Maps

DATE:  ca. 776 A.D.

AUTHOR:  Beatus of Liébana (died 798)

DESCRIPTION:  It is in Spain, in the early days of Muslim rule and in the heat of the Adoptionist controversy, that the original of the first important group or school of Christian European maps is to be found.  The eighth century Spanish priest Beatus of Liébana has been identified as the draughtsman of that plan which is the common source of the maps of St. Sever, Turin, Ashburnham, and eleven others of the earlier Middle Ages, executed at various times between at least the 10th and 13th centuries, but all depending on a Spanish-Arabic prototype of the eighth century. This prototype, now lost, originally appeared anonymously as a feature of a richly illustrated work, The Commentary of the Apocalypse of St. John, which has been fixed by both internal and external criticism to a date in or near the year 776 A.D, then revised in 784 and again in 786. Therefore, the term “Beatus” identifies a particular medieval manuscript, generally of Spanish origin, that contains a collection of textual comments on the apocalypse of Saint John. The aim of the author, Beato of Liébana, was that of indoctrinating and educating the clergy, although, in some cases the manuscript was also used for certain rites and rituals.

The first version of the commentary was successively edited by the very hand of Beatus, as well as by later authors, each of whom contributed in creating different versions. There are 28 illuminated manuscripts that have been identified as having these characteristics and are, therefore, named “Beatus of Liébana” and are conserved in various libraries around the world.

Beatus created an important Christian cultural and religious focal point during the eighth century. He corresponded with Alcuin, and took part in the Adoptionist controversy, criticizing the views of Felix of Urgel and Elipandus of Toledo. As confessor to Adosinda, wife of Silo of Asturias, and as the master of Alcuin and Etherius of Osma, Beatus exercised wide influence. His Commentary was popular during the Middle Ages and survives in at least 34 manuscripts (usually called a beatus) from the 10th through the 16th centuries. At least 28 of those manuscripts contain illuminations. Though Beatus may have written his commentaries as a response to Adoptionism in Spain of the late 700’s, many believe that the book’s popularity in monasteries stemmed from the presence in Spain of Islam, which the Christian religious believed to represent the Antichrist. Not all of the manuscripts are complete, and some exist only in fragmentary form. Twenty-seven of these manuscripts are lavishly decorated in the Mozarabic, Romanesque, or Gothic style of illumination.

In 776, Beatus, a monk and probably abbot of Santo Martino in the Liébanese valley near the northern coast of Spain, finished compiling his Commentary on the Apocalypse. Beatus divided his Apocalypse into 68 “sections” (storiae) of a dozen or so verses. Each storia was followed immediately by its illustration, partially or totally filling a page or, unusually for medieval illumination, spread uninterruptedly across two pages. Each illustration was directly followed by an explanatio consisting of a series of interwoven exegetical passages that interpreted in allegorical and anagogical terms each of the verses or figures in the storia just presented.

Only one page of introduction of the entire Commentary text, which runs to a thousand pages in the modern edition, was actually composed by Beatus himself. The original authors of the interwoven exegetical passages were proudly identified in the Preface (I, 5-6) as Jerome, Augustine, Ambrose, Fulgentius, Gregory, Tyconius, Irenaeus, Apringius and Isidore. Of these, the North African Tyconius made by far the most important contribution. If Beatus was a “cut and paste” editor, he was an industrious and motivated one, driven, the historian John Williams believes, by the anticipation of the end of ordinary time. The arrival of Spanish era 838, that is, A.D. 800, would end the sixth of the millennial weeks and usher in the seventh and last, setting off the events prophesied in the Apocalypse.

A page from the Beatus Codex of San Andrés de Arroyo

The Commentary includes prefatory material unconnected with the Apocalypse. The Prologue to Book II, for example, begins with a discussion of Ecclesia et Synagoga in which the various ranks of the holy and dedicated are identified and described. Among them are the apostles. For this section Beatus appropriated De Apostolis from Isidore’s Etymologiae (Book VII, 9, 1-4; see #205). He then inserted a short portion of the De Ortu et Obitu Patrum sometimes attributed to Isidore:

These are the twelve disciples of Christ, preachers of the faith, mentors of the people. Although all carry out the same work, each one of them received a specific region to preach in: Peter Rome, Andrew Acaia, Thomas India, James Spain, John Asia, Matthew Macedonia, Philip Gaul, Bartholomew Licaonia, Simon Zelotes Egypt, James the brother of the Lord Jerusalem. Unlike the others, Paul did not receive a particular post, but was elected mentor and preacher to all the people. For as Peter and the others were apostles of the circumcised, Paul was for the gentiles. He also evangelized the seven churches and the three disciples.

Although Beatus acknowledges in the opening sentence that there were twelve apostles, he omitted Thaddeus (Jude) and Matthias, Judas’ replacement. These were included in his source, De Ortu et Obitu Patrum, however. Beatus concluded this section on the apostles on an allegorical note:

These are the twelve hours which through Christ illumine the day; these are the twelve gates of the heavenly Jerusalem through which happy life is entered. They constitute the first apostolic church, which we trust as a foundation stone firmly anchored in Christ. These are the twelve thrones which are to judge the twelve tribes of Israel. This is the church reaching throughout the world. This is the sacred and select seed, the royal priesthood disseminated throughout the world. They were few but select and from this small seed arose a great harvest. We have faith in and possess this Church, and that which was evangelized before these is not Christian but will endure perpetual anathema, excommunication, that is, perdition when the Lord comes. And this is shown to greater effect these grains of seed scattered in the field of the world which the prophets prepared-by the pictorial formula attached.

The Commentary also contains one of the oldest Christian world-maps, thought to represent a description given by Isidore of Seville in his Etymologies. Although the original manuscript and map have disappeared, beginning with the 10th century derivatives of the map survive in several of the extant manuscripts. For more information on Beatus of Liébana or on the Beatus Apocalypse manuscripts, see the studies carried out by John Williams, Mireille Mentre, José Camón Aznar, Wilhelm Neuss, Joaquín Yarza Luaces, Sandra Saenz-Lopez Perez, among others (see References).

The derivative Beatus maps were first presented in a magisterial and unsurpassed study by Konrad Miller in the first volume of his monumental six-volume survey of 1895-1898 of medieval mappaemundi. Miller concluded from the places mentioned on the Beatus maps that they descended from a fourth century Roman world map of the provinces and important cities of the Roman state, the type of map that provided the general model for medieval mappaemundi. That Beatus simply copied the essential form and content of an older world map of this type was in Miller words, “self evident”. The extant Beatus derivative maps were originally classified by Professor Miller into two main “Families” or “Branches”: those belonging to the Osma tradition and those of the Valcavado tradition.  Although a colophonic note (UOV) clearly dates the manuscript to 1086, Miller assigned it to 1203, a mistake that persisted in the cartographic literature for an astonishingly long time considering the number of studies generated by the Beatus Commentaries and surely reduced its potential status as a witness to map history. According to Miller, to the Osma family [Branch A] belong the examples of:

1060 A.D. (Paris I/St. Sever),

1050 (Paris III),

1086 (Osma), and

1189 (Lorvão);

Miller’s Branch B was represented by a Commentary written at the Leonese monastery of Tabara around 940 and now in the Morgan Library in New York (Morgan 644, a.k.a. Magius, Ashburnham, New York I), which contains the oldest surviving Beatus map. The following maps belong to Miller’s Valcavado tradition [Branch B]:

894/960 A.D. (Morgan 644/Ashburnham/New York I/Magius),

970 (Valladolid),

10th century (Urgel),

975, (Girona),

1047 (Madrid),

1109 (London/Silos),

1150 (Turin),

1175 (Manchester),

1180 (Navarra/Paris II), and, finally

1220 (Hueglas/New York II).

Miller was certain that both archetypes went back to Beatus himself. His reconstruction was anomalous; for while Miller made it clear he considered the Osma map to be the best witness in the number of place names, its oval shape and its acknowledgment through the apostles’ heads and names of the very purpose of the map - he believed that Morgan 644/Ashburnham/New York I, the oldest Commentary exhibiting the shape of most of the extant maps, reflected the first draft of Beatus’ maps. Miller thought the Osma map was the fair copy, even though Morgan 644 had only 96 place names compared with Osma’s 120 and a rectangular shape rather than an oval one. According to Miller, the parting of these two families probably took place in the ninth century, and each appears to have been immediately derived from certain lost intermediates of the 10th century, such as the two executed in whole or part by Emeterius of Valcavado between 968 and 978 A.D.  In all, the best known are the large, usually oval/round, maps that can be traced back to the prototype.

Beatus Chronology based upon Konrad Miller

In 1931, Wilhelm Neuss concurred with Miller’s assessment that a Late Roman model lay behind the Beatus maps. In his monograph on the illustrations in the Beatus Commentaries, Neuss also constructed a family tree based on independent collations of the Commentary texts and their pictures. Differences in the texts allowed Neuss to construct a family tree of two basic branches, with Branch I favored as the closer to the original recension. Branch II was in turn divided into sub-branches, IIa and IIb. A separate analysis of the illustrations led him to an identical stemma. Some copies of the Commentary Neuss identified as belonging to an earlier stage in the evolution of the illustrated text and as closer to the archetype, which he dated, arbitrarily, to c.785. Other copies, including Morgan 644, had suffered pictorial attrition in the evolution of the family tree. Except for one detail, Henry A. Sanders, Neuss’ contemporary, arrived at the same groupings during an independent analysis of the text. Neuss favored a common descent of Branches IIa and IIb from an archetype, whereas Sanders implied that IIb descended through IIa, a position that John Williams has also adopted.

Family Tree of Beatus Commentaries as proposed by Wilhelm Neuss in his Die Apokalypse.

The Arabic numeral above and below, at the start of each line, is the number assigned by John Williams in his The Illustrated Beatus. The folios indicate the location of the map in the codex. The bracketed Roman numeral at the end of each entry represents the branch of Nuess’ family tree.

2. Morgan 644, a.k.a. Ashburnham, New York I, Magius, (c. 940-945) Pierpont Morgan Library, New York, ff.33v-34 [IIa], 387x285mm.

4. Valladolid, (970) Biblioteca de la Universitaria, Ms 433, ff.36v-37 [IIa], 350x240mm.

6. Girona, (975) Museo de la Catedral de Girona, Num. Inv. 7 (11), ff.54v-55 [lIb], 400x260mm.

8. Seu d'Urgell, (last quarter 10th century) Museo Diocesano de La Seu d’Urgell, Num. Inv. 501, ff.VIv-VII [IIa], 402x265mm.

11. Madrid, (1047) Biblioteca Nacional, Cod. Vitrina 14-2, ff.63v-64 [IIa], 350x280mm.

13. Saint-Sever, a.k.a. Paris I (c.1050) Paris, Bibliotheque Nationale, MS Lat 8878, ff.45bisv-45ter [text I, map II], 365x280mm.

14. Burgo de Osma, (1086) Archivo de la Catedral Cod. 1, ff.34v-35 [I], 360x225mm.

15. Turin, (1150) Biblioteca Nazionale Universitaria, Sgn. I.II.1 ff.45v-46 [IIb], 372x296mm.

16. London, a.k.a. Silos, (1109) British Library, Add. Ms 11695, ff.39v-40 [IIa], 378x235mm.

20. Rylands, a.k.a. Manchester, (c.1175) John Rylands University Library, MS Lat. 8, ff.43v-44 [IIb], 454x326mm.

22. Lorvão (1189) Lisbon, Arquivo Nacional da Torre do Tombo, ff.34bisv [I], 345x245mm.

23. Navarra, a.k.a. Paris II, (1180) Bibliotheque Nationale, Paris, MS nouv. acq. lat. 1366, ff.24v-25  [I], 350x230mm.

24. Heulgas, a.k.a. New York II, (1220) Pierpont Morgan Library, New York, M. 429, ff.31v-32 [IIb]a, 530x340mm.

25. San Andres del Arroyo, a.k.a. Paris III (1248) Bibliotheque Nationale, Paris, MS nouv. acq. lat. 2290, ff.13v-14 [IIb], 440x305mm.

In addition, two maps of the Beatus type appear outside of Beatus Commentaries:

Oña (late 12th century) Biblioteca Ambrosiana, Milan MS. F sup. 105, ff.71V-72 b

A 12th century mural map at the Galician church of San Pedro de Rocas c


a. This map reflects a much earlier one, for it was copied from the Commentary (Madrid, Archivo Historico Nacional, Cod. 1097B) produced at Tabara in 970 by Maius’ pupil, Emeterius.


b. Luis Vazquez de Parga, ‘Un mapa desconocido de la serie de los “Beatos”’, in Actas del simposio para el estudio de los codices del’Comentario al Apocalipsis’ de Beato de Liébana (see text note 37) I*: 273-278. A manuscript from San Salvador de Ona. The map closely resembles that of Burgo de Osma, but lacks a Beatus companion Commentary text.


c. Serafin Moralejo Alvarez, ‘Mapa de la diaspora apost6lica en San Pedro de Rocas: Notas para su interpretaci6n y filiacion en la tradicion cartografica de los “Beatos”’, Compostellanum 34: 3-4 (1986): 315-40. It resembles the Osma map but has deteriorated badly.

From the philological analysis, some Beatus Commentaries were assumed to be better witnesses than others to earlier stages of the tradition. Both Neuss and Sanders agreed that the Commentary written at Saint-Sever in the middle of the 11th century employed the primitive text. Since the pictures in the Saint-Sever copy could be considered the richest in detail and artistic adroitness, Neuss concluded this was the closest to a Beatus archetype. Assured by his reading of the textual and pictorial evidence, Neuss did not so much demonstrate as assert that the Saint-Sever map was also the best reflection of the original Beatus map, saying: “not only is it by far the most complete, but it is the most antique in every respect”.

With 270 place names, the Saint-Sever map has more than twice as many names as the Osma map and three times as many as the Morgan 644. However, a large number of the Saint-Sever names are post-Antique, and many of these additions (50 places, 10 rivers, six provinces) are concentrated in a hugely exaggerated Gascony, a provincial name that originated only in the 10th century. The map would appear to reflect an overriding interest in “the political organization of the Gascon principality at the beginning of the abbatial rule of Gregory [Muntaner]”, the well-connected abbot for whom it was copied. It was the medieval nature of its nomenclature that had led Konrad Miller to reject the Saint-Sever Commentary as the best witness to the archetype.

Reading Neuss closely, we find he was in fact admitting, in a way never acknowledged by him for the Apocalyptic illustrations, a creative factor when it came to the crafting of the Saint-Sever map:

Outside of a few entries, easily recognized, especially Saint-Sever itself and some neighboring regions, it [the map] has nothing in it that a modern recreation would not have displayed. The other maps show an ever-widening departure from the antique archetype. The painter of Saint-Sever went, as it were, in the opposite direction and drew a map that was the key to all the others ... Thus [the map] provides still another indication that Saint-Sever is closest to the archetype.

This is a slippery conclusion: the Saint-Sever map is the best witness to the archetype because the painter worked to make it so! According to Williams, recognition of the possibility of creative revision, of mere “up-dating” or renewal prompted by changing standards is of fundamental importance for the history of the Beatus maps.

Gonzalo Menendez Pidal was the next to undertake a consideration of the origin of the Beatus map in a 1954 article. His somewhat arcane title, Mozarabes y asturianos en la cultura de la alta edad media, reflected the fact that it belonged to the campaign, spearheaded by the historian Claudio Sanchez Albornoz, to assign greater weight to the native as opposed to the Muslim contributions to early medieval Hispanic culture. As a result Menendez Pidal emphasized the Isidoran map tradition and shifted the discussion of the model used by Beatus from a vaguely designated Late Roman one to an Isidoran one, where it has remained ever since. 

In 1976 Peter Klein meticulously reworked Neuss’ analysis of the descent of the Beatus Commentary and showed it to be fundamentally flawed. Klein dealt only briefly with the map, but he effectively eliminated the Saint-Sever map as a contender for privileged witness to the archetype. Where Neuss had proposed a single archetype of around 785, which was as pictorially complete as the richest copies known today - that is, like Saint Sever - and which in its descent had been subjected to deterioration in copies like Morgan 644, Klein demonstrated a pictorial evolution that illustrated the growth or increase in size by gradual external addition. Even if the text of the Saint-Sever copy originated in the primitive stage, its iconographic richness, including the map, had to be in part owed to developments no older than the beginning of the 10th century.

Regardless of this controversy, the original prototype map seems to have had a special purpose beyond simply geographical. As a theologian, Beatus no doubt, considered his map as primarily illustrative of the Old and New Testaments, and of the spread of the Catholic Faith.  In addition to the scriptures, however, he also seems to have relied upon two more temporal authorities: St. Isidore of Seville (#205) and a Roman province-map that bore some resemblance to the Peutinger Table (Book I, #120).  From his fellow Spanish theologian Isidore, Beatus extracts almost verbatim most of the longer inscriptions or legenda; although it must also not be forgotten that Isidore himself derived the material for much of his geographical dissertation from the cosmographies of the later Roman period.  On the other hand, no earlier source is known for the apostolic pictures used by Beatus, as far as embodiment on a mappamundi is concerned; thus this detail may well be a refinement supplied by Beatus himself.  As to the Roman province-map, it is from this map that Beatus and his copyists are thought to have derived most of their ‘secular geography’.  The Caspian Sea, the Alexandrian Pharos, the Nile inscription, and the desert where the Children of Israel wandered for forty years, as we have them in Beatus cartography (especially the St. Sever map), are closely parallel to the representations of the Peutinger Table.

General Stemma for the Large Beatus Maps. This genealogy shows the lineage of the extant Beatus manuscripts containing full-page maps. Two main versions or drafts are shown, from the 8th and 9th/10th centuries respectively, each with two recensions. Adapted from Peter K. Klein, Der altere Beatus-Kodex Vitro 14-1 der Biblioteca Nacional zu Madrid: Studien zur Beatus-Illustration und der spanischen Buchmalerei des 10. Jahrhunderts (Hildesheim: Georg alms, 1976).

According to the historian John Williams, Beatus of Liébana’s Commentary on the Apocalypse (Spain, 776) also drew heavily on Tyconius’ Commentary (North Africa, late fourth century), which is a probable source for the world map Beatus used to show the apostolic dissemination of the Faith. A map based on Orosius’ Historiarum adversus paganos libri VII [Seven Books of History against the Pagans] is another possible model for Beatus’ map, and it would have differed only slightly from one following Isidore of Seville’s Etymologiae. Surviving maps in Beatus Commentaries incorporate features from both sources. One notable feature from Isidore is an interior ocean separating a fourth continent from the usual three parts of the oikoumene [known world]. Contrary to common interpretation, this fourth part represents not an anti-ecumenical hemisphere but the southernmost zone of the inhabited world. Although the map in the Beatus codex now in Burgo de Osma (dated 1086) is usually recognized as the best witness to the original Beatus map, the map in Morgan 644 (dated 940, a.k.a. the Ashburnham, Magius, New York I map) may provide a more reliable guide.

According to the historians Beazley and Miller, the relationship between these two works is a key to all satisfactory study of the Beatus’ Spanish designs.  This relationship is further demonstrated in many other details such as in the names of the peoples, cities, hills, and rivers of various countries, and in the Indian, Syrian and African legends.  In Gaul, not only are the same provinces named and the same divisions made, but the more striking omissions of the Peutinger Table also occur in Beatus copies such as the St. Sever.  Of the more than 130 names of towns that appear in the Beatus maps, more than ninety of them agree with the Peutinger Table, and among these ninety parallels, all except two of the important places are marked by pictures (usually houses); on the other hand, the great vignettes at Rome, Constantinople and Antioch, as seen in the Table, if existing in the works of Beatus, have undergone transformation and reduction to a much lesser level.  However, the Beatus designs have nothing similar to the Roman Itineraries, or to the station-, distance- and road-markings of the Peutinger Table.  Nor, of course, can the latter’s 600 references to pagan temples and worships be found in these designs of the 10th and subsequent centuries.  But, in spite of whatever differences exist, it appears that in the various works derived from the Spanish priest of Liébana and Valcavado, there can be found a medieval reflection of one or more cartographical works of the Old [Roman] Empire, free from all additions of the Crusading period, and of inestimable value as a link between the ancient and medieval worlds.

Additional significance of the 14 copies of Beatus that can a world map lies in their relatively high antiquity.  Four of them certainly belong to the pre-Crusading period, namely, St. Sever (a.k.a. Paris I), Ashburnham, (a.k.a. Morgan 644, Magius, New York I),Valladolid and Madrid; and, both from their age and size, these plans are worthy of notice.  There is scarcely anything from Latin Christian cartography of so early a date; and the few specimens which carry back to a still older time are considered mere sketches by comparison, such as the Albi map of the eighth century A.D. (#206), and still earlier the sketches by Cosmas in the sixth century (#202).

From the close similarity among all members of the Beatus Family, we can therefore deduce with some certainty the character, not merely of the primitive copies or intermediates, but also of the original itself, as drawn by the “obscure hill-man and cave-dweller” in about 776 A.D.  Of these 14 Beatus derivatives, nine are in the shape of an oval, somewhat inclining to the oblong; the oldest one, Ashburnham (a.k.a. Morgan 644, Magius, New York I), is a right-angled rectangle; and four of them, the Turin, Navarra/Paris II, Ona, and San Andres de Arroyo, are circular.  What was the original form?  This question can only be answered by reviewing the conditions under which it was drawn.  All of the copies mentioned are drawn on two pages; each page gives half of the map, or displays half of the known world; and perhaps the oblong shape so often to be used is due merely to the copyist lengthening the two halves of the circle in order to fill-up his space and give the work more room.  The height of the map is, of course, the height of the accompanying manuscript in all instances, thus supporting the theory that the elliptical form was circumstantial.  The comparatively short, upright axis, from top to bottom of the single page, represents the longitude, or east-to-west prolongation of the earth; while the breadth, the comparatively long horizontal axis reaching across the two pages, represents the latitude, or north-to-south extension of the world.  But neither in classical antiquity, nor in the Middle Ages, do we meet with any geographer who believes the latitudinal extension of the oikoumene to be greater than the longitudinal.  If this were so, then the very terms of longitude and latitude themselves would have been disputed; but, on the contrary, they were always accepted.  Hence, it will not do to use the Beatus maps as proof that the ancient Orbis picti, and especially the world map of Agrippa (#118 in Book I), were oblong or elliptical.

It is probable that on the original Beatus map both the Four Sacred Rivers and the Ancestors of Mankind were depicted, as on the ‘family’ of maps that are associated with Henry of Mainz (#215, #225, #226).  The four rivers of Paradise are a reference to Genesis ii, 11-14.  The first three rivers are usually identified with the Indus or Ganges, the Nile, and the Tigris.  Also, in the Beatus maps, there is no clear evidence of a dependence on the T-O design so dominant during this period of cartography.  The horizontal line dividing Europe from Asia is normally crowded up towards the top, and thus deflected from the actual middle (only on the Osma map is there an exception).

Beatus seems to have followed Isidore in his limiting of Africa to the North of the Equator, this was also the practice of many of the classical geographers such as Cicero, Pliny and Mela.  Although he never displayed it on his maps, Isidore conceded the probable existence of the southern Antipodes, and, based upon a single sentence or two from his pen, all of the Beatus copies (except the Navarra/Paris II map of 1180), portrayed an unknown continent south of Africa and the Indian Ocean.  Even the Navarra/Paris II map, however, gives a relic of the ‘Australian Continent’ by indicating, in a corner, the Skiapod, a shadow-footed monster whom the Osma map of 1086 shows in the ‘Southern Land’; this last was doubtless the original position.

As to the appearance of the circumambient fish and boats, these fish occur in every Beatus derivative except the Turin map; the boats are found on the following copies: St. Sever, Ashburnham, Valladolid, Girona, and the Navarra/Paris II maps.

In the original Beatus design, as in most medieval maps, the Mare Rubrum [Red Sea] appears to have been colored according to its name; but on the Navarra/Paris II map of 1180, the Valladolid, and the Ashburnham, this tint is confined in a more modern sense to just the Arabian and Persian Gulfs; both of these gulfs on the Ashburnham and Valladolid examples are depicted rather like the symbols used for denoting mountains than seas, in red color; in the Madrid map only the Arabian Gulf is red in color; on the London copy the mountain-like appearance is even more pronounced; on the Girona, Turin and Navarra/Paris II maps, both gulfs are tinted with the traditional hue of the sea.  On every Beatus map Professor Miller recognizes a trace of the original legends in this area.  The Black Sea, the Sea of Azov and the Caspian Sea occur on the St. Sever and on the Osma maps, but are wanting on all maps of the Miller-defined Valcavado group.  The distortion of the Mediterranean Sea fares no better on these maps than on the other cartographical works of the early Middle Ages.

The rivers of the world are more realistically portrayed on the best copies of Beatus, the St. Sever and the Osma; on the other examples, and especially on the Navarra/Paris II map of 1180, the representation of the streams may sometimes be used for restoring the probable contents of the original Beatus map.  Thus, apparently the original map contained no Spanish rivers, but marked the Rhine, the Rhone, the Danube, the Euphrates, the Tigris, the Jordan, the Nile with its delta, and certain effluent of the Caspian.

Where all of the copies agree, we may suppose that we are dealing with material from the original Beatus prototype; and, fortunately, the coincidences between and among all of the derivatives are so numerous that we can, from these alone, form a pretty detailed picture of the fundamental draft. Therefore, to summarize, the original Beatus design may have presented a schematically drawn, basically oval map designed to promote the spread of the Christian Faith and may have contained such features as a Southern Land or Antipodes, a circumfluent ocean filled with fish and possibly boats, colored seas with some being depicted as mountains, at least the seven aforementioned rivers, and an ample distribution of pictorial displays and textual legends.  The text of some of the longer legends gives a fairly good idea of the anthropological outlook of Beatus, his copyists and the medieval European.  The following is such an example:

Albania, so called from the whiteness of its people, and color of their hair, extended from the east, close to the Caspian Sea and the shore of the Northern Ocean [into which the Caspian was believed to flow] to the Maeotid Lakes [Sea of Azov], through desert regions where the dogs were so strong and fierce that they could kill lions. Hyrcania, so called from the Hyrcanian Wood [a confusion with the Hercinia Sylva of Germany] which lay ‘under Scythia’, was full of tigers, panthers, and pards. Many races lived here and in Scythia, among them cannibals and blood-drinkers. Scythia, stretching from the extreme east and the Seric Ocean, to the Caspian Sea (at the setting sun) and southward to the ridge of Caucasus, abounded in gold and gems, in the best emeralds and in the most pure crystal; but all of these treasures were guarded by Gryphons, and no man could approach thereunto. Armenia, between the Taurus and the Caucasus, and between Cappadocia and the Caspian, was divided into two parts, the ‘Greater’ and the ‘less’, and contained the source of the Tigris. Arabia, the land of incense and perfumes, of myrrh and cinnamon, of the phoenix and sardonyx, was also called Saba, from the son of Chus. The Dead Sea, so named because it produced nothing living, and received nothing from the race of living things, was in length 780 stadia and in breadth 150 [this attempt at measurement is a very unusual feature on a medieval map, and shows a curious, if inaccurate, precision, or spirit of inquiry (the figures that are given are twice too great). Beatus also gives measurements, in Roman miles, for the islands of Britain, Corsica, Sardinia and Taprobane.] India, containing many peoples and tongues, men of dark color, great elephants and precious products, such as gems, ivory, aromatics, ebony, cinnamon and pepper, was also famous for its parrots, dragons and its one-horned beasts [rhinoceros]. It was amazingly fertile, with crops twice a year; and among its gems were diamonds, pearls, burning carbuncles, and beryls; it also possessed mountains stored with gold, and guarded by dragons and monstrous men. Among its islands were Chryse [the Malay Peninsula] and Argyre, the isles of gold and silver and Taprobane, which lay far to the south, was divided by a river, was only in part inhabited by men, had ten cities and was full of jewels and elephants. Ethiopia, stretching to the borders of Egypt, abounded in races of diverse color and monstrous form. It possessed multitudes of wild beasts and serpents, precious stones, cinnamon and balsam. The Nile was said by some authors to rise far from Mount Atlas, and thereafter to be speedily lost in the sands. But soon it emerged from the desert, poured itself out into a vast lake, and thence flowed to the Eastern Ocean, through Ethiopia. Here, again, bending to the left, it descended upon Egypt.

Of these legends on the Beatus maps, most of them are to be found in the writings of Isidore, but some have, ultimately, far more ancient origin.  Thus the notice of Parthia plainly refers to a time before the Persian revival of 226 A.D.; while the dimensions of the Dead Sea and Lake Gennesaret, in stadia, also prove a considerable antiquity, perhaps back to a source at the time of Pliny.  The legends referring to the Hellespont and the Bosphorus correspond, in substance, with the descriptions of these areas found in the writings of not only Pliny, but also Mela and Solinus; the measurements of the greater islands (i.e., Britain), in Roman miles, seem to be reminiscent of an imperial Itinerary.

A general stemma [genealogy] for the large Beatus maps which shows the lineage of the extant Beatus manuscripts containing full-page maps is provided in Harley’s History of Cartography, Volume One (adapted from Klein). Below is a sample of the various shapes that the Beatus derivatives have taken.

The Shape of the First Beatus Map

As already noted, Miller designated the map in the Commentary of Burgo de Osma as the best record of the first Beatus map, or at least, of the fair copy made after the first sketch based on a Late Roman model, a sketch which is also reflected by the Morgan 644/Ashburnham copy. Subsequent refinements of Beatus scholarship have not undermined the authority Miller assigned to the Osma map but rather have enhanced it. Neuss’ and Sander’s analysis showed that the Osma text belonged to a primitive Branch I edition, which grants the map a greater claim to represent an original stage of Beatus imagery. The Osma map would have to share this privilege theoretically at least, with the other Branch I maps. However, despite the fact that there are ten Branch I Commentaries, evidence for the Branch I maps is weaker than this figure suggests for the map including Osma, and the character of the other two, Lorvão and Navarra/Paris II (Nouv. acq. Lat. 1366) reduces their value in the argument.

Only the southern half of the map in the Lorvão Commentary of 1189 survives. The other Branch I map, in a late 12th century Commentary in Paris, is anomalous. Its maker rotated the standard map a quarter circle counterclockwise, so that south is at the top. Nor does correcting the rotation resolve the map into a standard type. Not only are Europe and Africa separated by water, as expected, but Asia, too, has been bisected. However, it is the Red Sea, not the Mediterranean, that falls between Europe and Africa, the Mediterranean having been displaced to become the body of water between Asia and Europe along a north-south axis. Paradise would need yet another quarter rotation to bring it to its proper location. Like the portion of the Osma map shown below, however, this Paradise has four rivers rather than Adam and Eve who stand in the Paradise of Branch II maps. A Skiopod, the ‘Shadow-Footed Man’ whose leg is lifted as a parasol, reclines near Ethiopia as he does on the Osma map, but the heads of the apostles that distinguish the Lorvão and Osma maps are absent.

The evidence for the appearance of Branch I maps, however, can be supplemented by two maps from outside the Beatus context. One is the late 12th century map in a manuscript from Oña; the other is the badly deteriorated 12th century mural map in San Pedro de Rocas in southern Galicia. The peculiarities of the Oña and San Pedro de Rocas maps, most particularly their oval shape and their identification of apostolic sites through heads, clearly link them to the Osma and Lorvão maps. They must be copies of otherwise unknown Branch I Commentaries, but the San Pedro de Rocas map’s fragmented condition and the Oña map’s augmented state textually, dilute their status as witnesses to the archetype.

The question of the relative archetypal authority of these details involves, theoretically at least, the relationship among these copies. The fact that the Osma map of 1086 is the oldest of the four related versions just cited, does not in itself make it the archetype for its kind. Serafin Moralejo proposed, in fact, an inverted relationship relative to chronology, with the Lorvão map of 1189 offering the best view of the original Branch I map, with the Oña map closer to the model for Osma than to Osma itself. From a comparison of the other illustrations in Lorvão and Osma, Williams has come to a similar conclusion. Differences between them rule out the possibility that Lorvão copied Osma. However, common eccentric details point to a close relationship. Since some of these details are ornamental formulas that are unlikely to be earlier than the 11th century, the model for Lorvão was probably not much older than the Burgo de Osma copy of 1086.

The comparison of the maps also suggests that Lorvão was based on a map related to Osma but representing a prior stage. As his Apocalyptic illustrations also demonstrate, the artist of the Lorvão Commentary was of inferior skill and his work is of scant interest and reliability. For example, the two triangular shapes that jut above the trans-African ocean are labeled mons duos but are clearly inspired by what are called Mare Rubrum on the Oña and Osma maps. They are, in fact, the Persian and Arabian Gulfs. The late dates and, even more, the peculiarly close relationship displayed by the surviving Branch I maps reduce their value as witnesses to the true nature of the original Beatus map of the eighth century, a weakness compounded by the failure of the other Branch I Commentaries to have preserved their maps. The result of an enquiry on the archetype is to be treated with caution.

        Two details of the Branch I maps - the oval shape and the use of the apostles’ heads to identify the areas of their missions - have reinforced the impression that the Osma map provides a good reflection of the archetypal Beatus map. These details contrast with the rectangular shape and absence of apostles on earlier Branch II maps, such as those in Morgan 644 and Girona. Miller took the oval shape to correspond to the imagined Late Roman model. Yet, enigmatically, Miller also assumed that the first draft from the Roman model was rectangular, like Morgan 644, and other writers have also postulated a rectangular archetype.

However, the pattern of ovals and rectangles in the evolution of the Beatus tradition substantially qualifies the support that the Branch I maps’ testimony seems to give to an oval archetype. The preference in later periods for the oval shape is reflected in the Saint-Sever map maker’s modification in the middle of the 11th century of a Branch II (and therefore rectangular) model. Moreover, in the two cases where we know the 10th century model for a later map the rectangular shape of the original gave way to an oval in the copy. Since the Osma map was designed even later than the Saint-Sever, the grounds for imagining an oval shape for its archetype are not substantial.

The assumption that the heads of the apostles were a feature of the original Beatus map is also open to doubt. It is true they are present in the maps of the Branch I Commentaries of Lorvão and Burgo de Osma and in the Beatus-type map from Oña. Since the map was included to show the apostolic missions, the apostolic portraits have been seen as integral to the very purpose of the original map. It is not axiomatic, of course, that this purpose required portraits. They may well be a sign of updating, of employing reliquary-like busts evocative of cult sites, in a formal language that would have been unthinkable in an early map tradition.

In the case of St. James, who in the Osma map is enshrined in the church of Santiago de Compostela, the site chosen developed only in the course of the ninth century, that is, after the fashioning of the first Beatus map. Moreover, we have seen that Santiago was not included on the Branch IIa maps even in the 10th century. More to the point, since mere heads do not present the map maker with an impossible challenge, it may be asked why, if they were so integral, did they survive only in this discrete, relatively late, set of maps?

We can question other details. Is the remarkable Skiopod of the Osma map a descendent of the archetype? It is present on the late 12th century Navarra/Paris II map (Nouv. acq. lat. 1366), but its failure to appear on the more orthodox Lorvão and Oña maps suggests that it is the result of another updating. Once more the Etymologiae (XI, 3, 23) provides the inspiration:

This region because of the heat of the sun is not known to us. And it is said to be inhabited by the Skiopodes. They have one leg and marvelous speed. The Greeks call them Skiopodes because during the summer, falling on their backs, they shade themselves with their huge feet.

Neither the Oña nor the Lorvão map employs this text, however. What Oña does have, perhaps as a result of the wide dissemination of the Branch IIb maps, is the fourth part of the world caption from Book XIV, 5, 17, followed by Isidore’s description of the Red Sea (XIII, 17, 2-3). The Lorvão map has no figure of a Skiopod but does refer to an inhabitant in a legend not taken from the Etymologiae: In this region the sun is so fierce it is unknown to us and uninhabitable. It is said men with wide feet [pedes latos] shield themselves from the heat with a foot.

Two centuries earlier in the Isidoran maps now in the Real Academia de la Historia in Madrid, a zone comparable to the fourth part of the world, the southwest corner of Africa, was labeled terra de pedes latos. However, the surviving evidence and the history of imagery make it more likely that the ‘land of the wide feet’ reflected a vulgar transmission of the scientific commonplace that extreme climates produced monsters. In the 12th and following centuries comparable images of fantastic creatures would become commonplace; it is difficult to imagine that a Sciopod was designed much before that of the Osma map.
The popular tradition seems to have inspired Martinus, the artist of the Osma Commentary, to find a more ‘scientific’ and elegant way of including a ‘wide foot’, based on a search through the Etymologiae and a vivid plastic imagination, to perpetuate a tradition going back at least a century and a half if not still further. It resulted in an interpretation that, as Moralejo pointed out, conflated the original notion of Antipodes as a people, the soles of whose feet are opposite ours, with the need for the inhabitants of Ethiopia to find shelter from the fierce heat. Consistent with the argument offered above - that this fourth part of the world where the Skiopod resided had borrowed its character from its neighbor, Ethiopia - Martinus found his Skiopod in Isidore’s description of Ethiopia (Etymologiae XI, 3, 23). Perhaps he was encouraged to choose this monster because of Augustine’s explicit recognition in the City of God (XVI, 8) that Skiopods were descendants of Adam.

        One feature of the Osma map, a Paradise of four rivers radiating toward the four corners of a rectangle, contrasts conspicuously with the image of Paradise in the Branch II maps, where Adam and Eve stand beside a tree. The Oña map employs the same scheme as Osma, but Paris II (Nouv. acq. lat. 1366) has a circular Paradise with meandering rivers, and Lorvão a small half circle with no rivers. Thus the Branch I maps fail to agree on the emblem of Paradise but are unanimous in excluding Adam and Eve. Which type of Paradise - Adam and Eve or the four rivers-stems from the archetype? In favor of the four-river scheme is its consistency as a geographical feature with a map setting, Moreover, Isidoran maps of the so-called Y-O type seen in the 10th century copies described in monograph #205 represent Paradise as a four-river emblem, albeit descending rather than radiating. At the same time, the Osma map betrays in many places an intelligent, even erudite, designer, and the very consistency of the four-river scheme may have recommended its introduction to Martinus, its painter. In favor of the priority of the Adam and Eve type of Paradise is the fact that the apostles’ missions were the result of the transgression of Adam and Eve, who stand in shame on the Beatus maps.

One Branch I Commentary, now in the Escorial but produced about 1000 in San Millan de la Cogolla, offers indirect evidence in favor of Adam and Eve. Although mapless, at the point where the map would have been inserted is a full-page depiction of Adam and Eve flanking a tree, the type marking Paradise in the Branch II Commentaries. Since the Escorial Commentary undoubtedly was copied from a Branch I model and its Apocalyptic illustrations exhibit the Branch I pattern, it is tempting to conclude that the Adam and Eve who took the place of the map were inspired by a map in the model in which Adam and Eve marked Paradise. This witness is of special importance, for it is older than the Branch I maps and not part of the small family of Commentaries holding the surviving Branch I maps.

As far as the archetype is concerned, there is no compelling evidence that it was oval, that it depicted the heads of the apostles, or that it contained any of the features which distinguish the Osma map from the Morgan 644/Ashburnham map - or, to be more accurate, for the map that served as the model for Morgan 644/Ashburnham (both text and iconography indicate that Morgan 644/Ashburnham is not the archetype for its particular Branch II Family). The Saint-Sever map indicates what could happen to a map of the Branch II type when it was copied in the middle of the 11th century for a Commentary carried out for an ambitious and well-connected patron like Abbot Gregory Muntaner.

Sahagun, where the Beatus of Burgo de Osma was carried out only slightly later, was also privileged. One of the many signs of Leonese royal interest was the installation of an ambitious monk and scribe from Cluny as its abbot. More than any other painter of his time, Martinus revised the centuries-old Apocalyptic iconography of the Commentaries. Like the Saint-Sever, the Osma map, too, may be the result of a radical reworking. Indeed, Neuss’ conclusion about the Saint Sever map, that the painter consciously strove to recreate an archetype, is probably more appropriate for the Osma map. The creativeness of Martinus’ approach, measurable in his handling of the inherited Apocalyptic tradition in the Osma Beatus, makes it likely that the Branch IIa maps like Morgan 644/Ashburnham are a better reflection of the archetype. The very fact that Morgan 644/Ashburnham was carried out by a painter with minimal cartographic interests makes it all the more likely that what is present was received rather than invented.

Each of the surviving maps is spread across two folios, and is oriented with East at the top, with an image of the Garden of Eden there, represented either by the four rivers of Paradise or an image of Adam, Eve, and the serpent. The maps have artistic renderings of mountains and rivers, and in some cases, of cities.

The abundant sea life on most of these maps is remarkable, particularly in contrast with the complete absence of land animals. Of the 14 maps of this group, three do not include any fish or sea monsters, but all of the other maps include at least fish in the circumfluent ocean, and three of the maps display more exotic sea creatures.

In summary, we may be quite certain that the eighth-century archetype manuscript of Beatus’ Commentary included a map: the text contains a phrase introducing the map, and the latest study of the text of the Commentary indicates that this phrase has been in the book from the beginning. According to Chet Van Duzer the presence of sea creatures on all but three of the Beatus mappaemundi, and indeed their presence on the two Beatus maps which are thought to be artistically closest to the archetype, renders it very likely that the archetype contained sea creatures. Thus the Beatus mappaemundi permit us to extend the history of sea monsters on medieval maps back to the eighth century.

Also Van Duzer believes that the sea creatures on the Beatus mappaemundi serve the obvious function of helping the viewer to distinguish between land and sea, and also lend a dynamic character to the circumfluent ocean, making the watery edge of the world just as visually interesting as the land, and suggesting the wildness of distant regions. As the part of Beatus’ Commentary on the Apocalypse that the map illustrates makes no mention of sirens or other sea monsters, and as there is no consistency or chain of influence in the specific sea monsters depicted on the Beatus maps, we can conclude that the illuminators of each manuscript were able to choose which sea monsters to paint.

Location of Beatus Map Derivatives:

Ashburnham (Morgan 644, New York I, Magius), ca. 926, Pierpoint Morgan Library, MS. M644, fols. 33v-34, N.Y.

Seo de Urgel, 10th century, Archivo Diocesano, Codex 4, Catedral Seo de Urgel, Lérida, Spain,
Valcavado,  ca. 970, Biblioteca Universitaria, MS. 1789, fols. 36v-37, Valladolid, Spain.

Girona, ca. 975, Museo de la Catedral, MS. 10, Gerona, Spain.

Saint-Sever (Paris I), 1060, Bibliotheque Nationale, MS. 8878 (S. Lat. 1075), fol. 45,Paris, France

Burgo de Osma, 1086, Archivo Catedral Burgo de Osma, MS. 1, fols. 35v-36, Soria, Spain

London (Silos, Reyes Fernando y Sancha), 1047, Biblioteca Nacional, Vitr. 14.2, fols. 63v-64, Madrid, Spain.

Turin, 1150, Biblioteca Nazionale Universitaria, MS. I.II.1 (old D.V.39) fols. 38v-39, Torino Italy

London, Silos, 1109, British Library, Add. MS. 11695, fols. 39v-40, London, England

Lorvão, 1189, Arquivo Nacional de Torre do Tombo, Codex 160, Lisbon, Portugal.

Navarra (Paris II), 1180, Bibliotheque Nationale, NAL 1366, fols. 24v-25, Paris.

Manchester(Altamira), 1175, John Rylands Library, MS. Lat. 8, fols. 43v-44, Manchester, England

San Andres del Arroyo, 1220, Bibliotheque Nationale, NAL 2290, fols. 13v-14, Paris, France

Mapa de Oña, late 12th century, MSS F. 105 sup. Dela Biblioteca Ambrosiana, Milano, Italy

Huelgas (New York II), 1220, Pierpoint Morgan Library, MS. 429, fols. 31v-32, New York.


*Bagrow, L., History of Cartography, Plates XV, XVI.

*Beazley, C., The Dawn of Modern Geography, volume II, pp. 550-559; 591-604.

*Brown, L. A., The Story of Maps, p. 127.

*Brown. L. A., The World Encompassed, no. 12, plate III.

*Delumeau, Jean, History of Paradise: The Garden of Eden in Myth and Tradition, p. 59

*Destombes, M., Mappemonde, A.D. 1200-1500, #17.

*Edson, E., Mapping Time and Space, pp. 149-159.

*Edson, E., The World Map, 1300-1492, pp. 17, 20, 25.

*Garcia-Araez Ferrer, H., La Cartogrfia Medieval Y Los Mapamundis de los Beatos.

*Galichian, R., Countries South of the Caucasus in Medieval Maps: Armenia, Georgia and Azerbaijan,pp. 66-81.

*Hapgood, C., The Maps of Ancient Seakings, p. 5; Figure 1.

*Harley, J.B., The History of Cartography, Volume One, pp. 287, 302-303, 331, 343, 357, Plate 13.

*Harvey, P. D. A., Medieval Maps, Plate 17.

Klein, P., Der ältere Beatus-Kodex Vitr. 14-1 der Biblioteca Nacional zu Madrid: Studien zur Beatus-Illustration und der spanischen Buchmalerei des 10. Jahrhunderts (Hildesheim: Georg Olms, 1976).

*Menendex-Pidal, G., “Mozarabes y astrurianos en la cultura de la A;ta Edad Media en relacion especial con la historia de los conocimientos geograficos”, Bulletin de la Real Academia de la Historia, 134 (1954): 137-292.

*Miller, K. Mappamundi: Die altesten Weltkarten, Stuttgart, 1895-98.

*Landström, B., Bold Voyages and Great Explorers, p. 89.

*Moralejo, Serafin, World and Time in the Map of the Osma Beatus, Apocalipsis Beati 

   Liebanensis Burgi Oxomensis, I, pp. 145-174.

*Nebenzahl, K., Maps of the Holy Land, p. 26, Plate 6.

*Nordenskiöld, A. E., Facsimile Atlas, p. 33, Figure 17.

*Perez, Sandra Saenz-Lopez, The Beatus Maps, Siloé, arte y bibliofilia, 2014.

*Perez, Sandra Saenz-Lopez, “The Image of France in the Beatus Map of Saint-Seaver” Space in the Medieval West, Chapter 8, pp. 159-173.

*Raisz, E., General Cartography, p. 14.

*Scafi, A., Mapping Paradise, pp. 104-116; 122-3.

*Talbert, Richard J. A., Unger, R. W., Cartography in Antiquity


For specific examples of the various Beatus derivatives, see the following (CLICK on the image #):

       #207.1 Morgan 644, a.k.a. Ashburnham, New York I, Magius, (c.940) Pierpont Morgan Library, New York, ff.33v-34 [IIa]

207.2 Valladolid, (970) Biblioteca Universitaria, Ms 433, ff.36v-37 [IIa]

207.3 Girona, (975) Museo de la Catedral, Num. Inv. 7 (11), ff.54v-55 [lIb]

207.4 Seu d'Urgell, (last quarter 10th century) Museo Diocesano, Num. Inv. 501, ff.VIv-VII [IIa]

207.5 Madrid, (1047) Biblioteca Nacional, Cod. Vitrina 14-2, ff.63v-64 [IIa]

207.6 Saint-Sever, a.k.a. Paris I (c.1050) Paris, Bibliotheque Nationale, MS Lat 8878, ff.45bisv-45ter [text I, map II]

207.7 Burgo de Osma, (1086) Archivo de la CatedraL Cod. 1, ff.34v-35 [I]

207.8 Turin, (1150) Biblioteca Nazionale, Sgn. LILlo ff.45v-46 [IIb]

207.9 London, a.k.a. Silos, (1109) British Library, Add. Ms 11695, ff.39v-40 [IIa]

207.10 Manchester, (c.1175) J. Rylands University Library, MS Lat. 8, ff.43v-44 [IIb]

207.11 Lorvão (1189) Lisbon, Arquivo da Torre do Tombo, ff.34bisv [I]

207.12 Navarra, a.k.a. Paris II, (1180) Bibliotheque Nationale, Paris, MS n.a. Lat. 1366, ff.24v-25  [I]

207.13 Heulgas, a.k.a. New York II, (1220) Pierpont Morgan Library, New York, M. 429, ff.31v-32 [IIb]a

207.14 San Andres del Arroyo, a.k.a. Paris III (1220) Bibliotheque Nationale, Paris, MS n.a. Lat. 2290, ff.13v-14 [IIb

207.15 Mapa de Oña, late 12th century, MSS F. 105 sup. Dela Biblioteca Ambrosiana, Milano, Italy