TITLE: The Bell Map

DATE:  ca. 1450


DESCRIPTION: This example of medieval cartography, referenced in David Woodward’s landmark essay on mappaemundi, or maps of the world from the Middle Ages, published in 1987, and briefly described in Marcel Destombes’s important census of world maps made between 1200 and 1500, which appeared in French in 1964. Both descriptions refer to the James Ford Bell Library’s map as a “fragment,” a word that conjures images of brokenness, minuteness, and relative unimportance.

Bound in a rather large slip-case on a piece of vellum, or fine parchment, about as large as a legal-size sheet of paper (24 x 33 cm) and after more than five centuries, the surface of the map, itself aged to a dusky ochre, still dances with design and hue. Waterways are depicted as teal swatches, delicately textured by parallel curved lines that represent wave action; several men at the top and right-hand edges look quite dapper in short tunics and leggings of vermillion and royal blue; cities appear as white architectural constructs with towers, domes, and roofs of red, green, and even flashing silver.

It is indeed a fragment that has been trimmed on each side, with a resulting loss of image and text all around it. Scattered throughout are words — names of cities and kingdoms, brief descriptions of local inhabitants, tidbits of historical lore — and they seem to be randomly distributed, an appearance encouraged by the absence of any lines suggesting regional or national boundaries. These words and phrases, known as legends (from legere, the Latin verb ‘to read’) are cartographical “sound bites”. While words on modern maps generally are restricted to the names of topographical features, medieval mappaemundi commonly employed legends to impart historical, theological, sociological, and scientific data. Included in the map’s Latin words and phrases, is a reference to the great Alexandrian mathematician and astronomer Ptolemy in a paraphrase from his book, Geographia (ca. C.E. 160). Ptolemy’s work was little known in western Europe for some nine hundred years following two references to it in the mid-to-late sixth century, and has often been said by scholars to have exerted little or no influence before it was first printed in the 1470s. Yet it seems from the handwriting of its legends that the Bell Library’s map is nearly a generation older than that.

      The Bell map is unusual in its basic format. Unlike most medieval mappaemundi, which have East at the top, this one is oriented to the South, a characteristic of Arabic cartography, which was a sophisticated science during the European Middle Ages but little known in the West. The image of the map printed here (center) will be more familiar to viewers who turn it upside-down: the Mediterranean coastline is now much more easily recognizable, with the Iberian peninsula especially evident (near the upper-left edge of the upside-down page). Africa can be made out with its familiar western bulge and southern extension, coastline features that startled me even more than the reference to Ptolemy, since they have long been thought to be geographical facts little known in mid-15th century Europe. At that time even the most elaborate mappaemundi continued an ancient tradition of representing the entire terrestrial landmass as lying north of the equator. Certain geographical treatises even raised doubts about whether Africa could be considered one of the Earth’s parts — anachronistically and somewhat misleadingly translated “continents” — owing to what Europeans believed was its small size and negligible population.

      In dealing with a text from the Middle Ages — whether made up entirely of words or combined with design — scholars often owe nearly as large a debt to a later discoverer of it as to its medieval writer or artist. This map is a case in point. The anonymous person who devised and decorated it deserves our admiration, of course. But John Parker, the first curator of the Bell Library, has been a key player in the process of identifying and preserving it as an important pre-modern image of the world. From Parker’s succinct article printed in the cartographical journal Imago Mundi in 1965, as well as from correspondence in the University of Minnesota Library Archives, we learn that he was looking for rare books and manuscripts in the shops of European dealers in 1960, when he paid a call on the antiquarian August Laube in Zürich. The visit was not particularly fruitful until he asked Laube if he might look through the drawers of an old bureau. There, amid relatively unremarkable prints of birds and flowers, lay the fragment of a world map, which so piqued his interest that Laube agreed to send it to Minneapolis. After examining it more closely with his assistant, Vsevolod Slessarev, Parker reached an agreement with Laube in late October, and the map has since remained on a continent that was not part of its inventive cartographer’s wildest imagination. Unfortunately, Laube had no information about the work’s provenance, and after his death any records of the shop were lost or destroyed. In his article in Imago Mundi, however, Parker insightfully observed that this hitherto-unknown fragmentary representation of the world bore affinities to two other mappaemundi — known as the Walsperger (#245) and the Zeitz maps (#251)— and that it was therefore related to what the historian Dana Bennett Durand, in 1934, first called the “Vienna-Klosterneuburg Map Corpus”.

      Parker regarded the Bell Map to be a “planisphere.” The word is a technical term for the representation of a sphere on a plane surface; since there is no documentary evidence to indicate that any Western map made after the 7th century attempts to represent the world as anything but a sphere, the word may describe any mappamundi. But a “planisphere” also refers to a depiction of the Earth with the planets and stars that surround it in an orderly series of spherical spaces: the pre-Copernican, geocentric image of the universe. Parker’s eye was no doubt caught by the upper-right-hand corner, where the remnants of several rings — in blue-green, black, red, and yellow-brown — can be seen encompassing the large, innermost circle that represents the terrestrial landmass. This part of the Bell Map’s design is extremely important, for it indicates that the cartographer was drawing a picture both of Earth and of the heavens. In addition, by measuring the curvature of the inner-most arc described there, we can calculate that the diameter of the Earth as it was shown on the original, complete map was approximately 20 inches (500/510 mm). Several letters of a word appear at the upper-right extreme, suggesting that this was the outermost ring (as is the case on the analogous Walsperger map, #245). If this is true, the surrounding rings added a total of around 3 inches (90 mm) to the display, which means that the Bell Map in its pristine state measured at least two feet (590/600 mm) on each side (this does not allow for any marginal space). Since the current dimensions of the map fragment, which has a pronounced bulge at the lower-right, are approximately 13 x 9 inches (330/334 mm by 223/238 mm), what survives is approximately one-quarter of the cartographer’s actual production, although if we remove the decorative border with the celestial spheres from consideration, nearly 40% of the originally depicted Earth remains.

So we are looking at a multifaceted puzzle: a map whose age, condition, and fragmentary nature raised considerable roadblocks to anyone who wanted to read it, and whose legends, once they were deciphered, seemed to say unlikely things. Nagging questions are at the core of most scholarly projects, however, since these are all about turning puzzles into knowledge.

      Certainly the best known of medieval travel works is the one attributed to Sir John Mandeville, who, according to the book’s prologue, written in the first-person, accumulated extensive lore and experience between 1322 and 1356 as an English mercenary soldier in the armies of the Egyptian sultan, the Great Khan in China, and (according to some manuscripts) Prester John, while also visiting pilgrimage sites throughout the Holy Land and in India, where the Apostle Thomas was said to have been martyred. During the second half of the 19th century, when it was conclusively shown that the imperial Prester John had ruled in the medieval European imagination rather than over any actual territory, scholars began to question Mandeville’s claims and even his very existence. Today the knight has lost most of his credibility, but the work popularly known as Mandeville’s TravelsThe Book of John Mandeville is historically a more accurate title — nevertheless ranks as the medieval equivalent to a best-seller: it survives in at least 300 manuscript copies and in ten languages.

      The Latin versions of the Mandeville book are of particular interest. Although they were copied by separate scribes, both are found in a single manuscript (today bound as two volumes), compiled between 1418 and 1455, uniting ten works about journeys from Europe to Asia in what may be described as one of the West’s first anthologies of travel narratives. One text appears without the formal announcement, or rubric (because it was often written in red ink), by which scribes called attention to the start of a new book, even though all the other works in the manuscript began with this medieval convention. It is a first-person account by one Johannes Witte de Hese who claimed to have had some truly outlandish experiences over the course of around three years, beginning in May 1389, while voyaging on the Red Sea, down the east African coast, and throughout Asia.

      Another 15th century manuscript in the Bell Library is a relatively late, considerably extracted copy of the Imago mundi of Honorius Augustodunensis, written around the year 1110, which enjoyed a remarkably wide circulation for nearly four hundred years. Little is known about Honorius except that he was the Benedictine author of several treatises summarizing various branches of learning. His aim, as he put it in his preface to the Imago mundi, was to draw information from authoritative sources in order to instruct people who do not have many books, which in the early 12th century was nearly everybody. Some 175 copies of it produced before 1500 (fewer than ten are accompanied by any kind of a map) have been located to date, and similarly “condensed” copies of the text that connect the Bell manuscript to humanist schools in 15th century Italy, southern Germany, and Austria, and thus to the very astronomers, mathematicians, and cosmographers to whom Christopher Columbus turned when he was seeking information as he developed his plan to seek east Asia by sailing west from Iberia.

      Despite its possible significance in the history of early modern exploration, the Imago mundi is a quintessentially medieval text. Like almost every other European who undertook to write about the shape of the Earth during the Middle Ages, Honorius describes it as a sphere, likening it to an orb and a ball, in passages so matter-of-fact that it is clear he expects his audience to agree with him completely. He even offers a calculation of the Earth’s circumference, and while it is always somewhat difficult to establish exactly the modern equivalent of a medieval measurement, his reckoning (or that of his source) appears to be remarkably accurate. Equally typical of medieval geography is the problem Honorius has in establishing precise spatial relationships in his descriptions of land areas. In moving westward from India, for example, he makes quite crude transitions, beginning successive brief paragraphs with such phrases as: “Parthia is [the land] from the Indus River to the Tigris,” “Mesopotamia is [the land] from the Tigris River to the Euphrates,” and “Arabia is also in [Mesopotamia].” Such rudimentary discriminations are reflected on many mappaemundi, which were drawn without reference to determined — or even imagined — lines of latitude and longitude, and on which land areas often occupied space according to their importance to the cartographer or his milieu: thus, a monk drawing the world often allowed as much room for the Holy Land as he did for all of India, since the confined territory where David reigned and Jesus walked was hierarchically significant enough to warrant making it seem as large as the Far East. Moreover, misunderstood or wholly fantastical reports on areas of Asia and Africa led some geographers, including Honorius, to report the presence of sensational monsters, societies of bizarre humans, and other wonders in remote parts of the world, especially in regions about which nothing concrete was known. 

      In dealing with the often anonymous, frequently incomplete, sometimes anomalous texts of the Middle Ages, a manuscript is frequently its own best informant. So it is with the Bell Map, about which a few key facts may be established by examining evidence found in four legends or legend groups.

In central Europe, near the bottom of the map, the cartographer has placed the towns of michenpurk and n’purh (the first is a town in Romania and the second an abbreviated form of Nuremberg). The endings of these words mark an interesting departure from the mapmaker’s general practice of giving all place-names in Latin. Here he is clearly thinking in German, as also appears to be the case when he writes doratz (Durazzo), zeng (Senj), and jenff (Geneva) in legends 80, 86, and 124. In the first two cases, the spellings purk and purh (for “burg” and “berg”) reveal even more, since the substitution of p for b and k for g are typical features of German dialects spoken in Bavaria and Upper Austria. Several cities are similarly identified with the ending purk on the Walsperger map (#245), which we know was compiled in 1448 in the city of Constance by a Benedictine from Salzburg. The Bell Map’s cartographer would appear to have been from the same region (scribes freely adopted the spellings of their own dialects in copying vernacular texts), but whether it was produced in a secular or a monastic context remains a riddle. It may well have been part of the library — or the assembly room — of a cloister that was secularized by Napoleon, passing from hand to hand occasionally as a curiosity for some 150 years until it landed in Laube’s antiquarian shop in Zurich. This might have been the fate of the Walsperger map (#245), had it not been acquired by the medieval University of Heidelberg, whose distinguished library was appropriated by forces loyal to the pope during the early 17th century, the contents being removed to Rome — in a move that is still today regarded by some as a theft and by others as a rescue from certain destruction — to become a magnificent part of a relatively new collection of books known today as the Vatican Library.


      Although they are on opposite sides of the surviving fragment, Legends 71 and 149 are linked in that each seems to offer information about the Bell Map’s date of composition. The second of these refers to the island Canaria as having been “recently discovered,” which sounds as if not much time had passed since the conquest of the Canaries in 1402 by Jean de Béthencourt and Gadifer de la Salle (colonization began in earnest three years later). This Atlantic campaign was glorified in Le Canarien, a popular poem of the period. But the best scribes produce good copies of a text, and this phrase may have been written decades before the Bell mapmaker treated it as if it were news. Legend 71, meanwhile, calls Constantinople the “capital city of the Greeks,” possible evidence that its capture by the Turks in 1453 had not yet occurred. Maps are important political tools, however, and toponyms have a singular power (as the history of St. Petersburg proves). Many Europeans refused to believe that the Turks would long control the city they re-named Istanbul and so this legend is not solid proof that it was written in the first half of the 15th century. The best evidence for a date is the handwriting, since calligraphic styles changed markedly during this period, and the Bell Map appears to be an excellent example of a German hand from around 1440/1460.

      Establishing where and when a medieval text originated is crucial to any attempt to learn to read it with an appropriate sense of context. Equally important is knowing something about the intellectual community that produced it. The Bell Map’s pedigree is quite remarkable, as is indicated by Legend 6, in which we learn that Ptolemy estimated the Earth’s landmass to extend for 180 degrees, or halfway around its spherical form, which was divided into 360 degrees. The detail comes from the polymathic Alexandrian’s great book Geographia, which was not generally known in western Europe after Greek became a lost language there during the fifth and sixth centuries. Jacopo d’Angelo completed his Latin translation in the first decade of the 15th century, but the work only gradually caught the attention of scholars because of its technical language and sophisticated concepts.

      Ptolemy’s ideas had special resonance, however, within a loosely connected group of monks, attached to the great Augustinian foundation at Klosterneuburg but also resident in Benedictine monasteries from eastern Austria to Bavaria, together with professors at universities from Vienna to Bamberg. They were especially creative and productive between 1420 and 1440. As Dana Bennett Durand first pointed out during the mid-1930s, these scholars developed technologies of measurement and theories of spatial representation based on mathematical and astronomical data, and in so doing revolutionized the way that geographies were written and maps were drawn. They were especially intrigued by Ptolemy’s argument that places in the “real” world can and ought to be located on a map by means of calculations. They were also amazed by his tables of coordinates fixing such places — the Geographia contains some 8,000 such “locations” — although they were not so awestruck that they hesitated to check and correct the master. Durand argued that members of what he called the Vienna-Klosterneuburg Circle had, by 1425, used a coordinate system to construct a world map (and at least two copies), which he attempted to “reconstruct” based on the principal evidence for its existence: coordinate tables found in a manuscript in the Bavarian State Library at Munich (which arrived there after the secularization of the Benedictine monastery of St. Emmeram in Regensburg). Durand’s assertions in this regard, which have gone unchallenged for half a century, may be subject to severe question, as Patrick Gautier Dalché has recently indicated. It cannot be denied, however, that these coordinate tables are related to two mappaemundi noted above: the Walsperger map of 1448 (#245) and the Zeitz map of ca. 1470 (#251). The Bell fragment is clearly a third member of this cartographical family, in its original state slightly larger than the Walsperger map, more interesting than its siblings for being richly illuminated but like them in the use of dots to show exact sittings of cities. (A scale for calculating distances is found at the bottom of the Walsperger map [it is more decorative than trustworthy].) The Munich manuscript that Durand studied contains descriptive material, lists of toponyms, and tables that record legends and their coordinates. These data prove that precision applied equally to the location of individual cities and to the “sprawl” of rivers and mountain ranges; they also contain specific instructions to enable an artist to render people and other design features. Thus, the compilers of the Bell Map and the other members of its cartographical family sought to find a scientific way — at a time when the printing press was not yet known or fully understood — to produce copies of a world map that consistently indicated the location of places with a precision unavailable in language to Honorius Augustodunensis and in design to the artists of other mappaemundi.

      Despite how “modern” the Bell Map may seem to be in this respect, it is also an extremely traditional medieval construct, as a fourth cluster of its notations indicates. Legends 8 and 10 locate the Isle of the Dead, an inheritance of classical mythology; Legend 13 puts sun-worshippers literally in their place; Legends 16, 17, 19, 20, and 22, respectively, tell us where to find people whose feet are so big that they may be used as umbrellas, as well as humans with goats’ heads and foxes’ tails and lions’ heads. To understand the ominous information in Legend 52 we need to look at a similar notation on the Walsperger map, which also tells of a nine-headed woman, making it clear that six of her heads are snakes. These resemble the wondrous — even monstrous — races of humans that occupy chapters in Honorius’s Imago mundi or (sometimes but not always) the margins of other mappaemundi.

      In the chart in Appendix 2 Mr. Westrem has tried to distinguish among land areas on the Bell Map and to identify which features of those areas the cartographer saw fit to include. Westrem includes Egypt and Mediterranean Islands as separate entries because their geographical status was by no means clear: most medieval writers put Egypt in Asia, but his doing so here would have contradicted modern convention. It is important to note that while some two-thirds of the Bell Map fragment’s area shows Africa, Europe has well over half the total number of Legends, almost all of which are names of cities or regions. One in four African Legends describes a people or an event, and the nature of either is almost always exotic. Egypt is a kind of median territory, with sites that include both Cairo (Legend 44, shown with great black towers topped by crescents) and the Temple of Ammon (Legend 28, a place famous in medieval Europe from the very popular set of stories, in prose and poetry, associated with Alexander the Great, one of the remarkable lost literary subjects of European literature). Legend 38 shows where Pharaoh drowned pursuing the Israelites, and below it the cartographer has drawn in the pathway of the exodus. History as recorded in the Bible, the matter of legends, and the stuff of romances (such as Herzog Ernst, a popular Middle High German poem of adventure on the high seas) combine with scientific calculations and a “modern” sense of how to represent space in a round world on a flat surface.

      The Bell Map’s lessons are testimony to its status as a rare example of innovative medieval cartography. If some of its legends are, to modern readers, rather silly — if ignorance about Africa led its maker into flights of fancy or to seek information from a book of what today seems an inappropriate genre — let that just be testimony to the fact that, even more than Nature, a cartographer abhors a vacuum.

Location: The James Ford Bell Library, University of Minnesota


Westrem, S.D., “Learning from Legends on the James Ford Bell Library Mappamundi”, The James Ford Bell Lectures, No. 37.


This transcription of the Bell Map’s Legends begins at the upper-right (southwest) corner and then moves gradually from top to bottom (south to north). The upper (southern) part of the map — some two-thirds of its area and almost entirely given over to modern-day Africa — has been divided into four horizontal bands, within which Legends (numbers 1-57) are recorded moving left to right (east to west). The lower (northern) third of the map — some of Asia Minor and around half of Europe — has been divided into eight vertical columns, within which Legends (numbers 58-152) are recorded moving from top to bottom (south- to-north). The following symbols or variations in typography should be noted.

|designates the edge of the map

#designates an illegible letter

_designates a minim (a basic stroke in medieval handwriting, used by the Bell Map’s scribe in forming all or part of the letters g, i, j, m, n, p, u, v

[ ]designates editorially supplied letter(s)

*designates a legend that is adjacent to an illustration (including an architectural design)

Ddesignates a reading in Durand’s Appendix 14 used to suggest/identify a Bell Map Legend

Wdesignates a reading on the Walsperger map used to suggest/identify a Bell Map Legend

Zdesignates a reading on the Zeitz map used to suggest/identify a Bell Map Legend boldface type identifies words written in red ink letters in italics are editorial expansions of scribal abbreviations