Modern Medieval Map Myths: The Flat World, Ancient Sea-Kings, and Dragons
By Michael Livingston
10 June 2002
I want to do a couple of things in this article. The primary purpose, as the title suggests, is to expose three modern myths about medieval maps. I am relatively certain that at some point in your life you were told, for instance, that Columbus proved that the Earth was round. I feel confident in this assumption only because I was taught the very same thing even though there isn't a shred of truth to the statement. No one, outside of a couple theologians (that only modern scholars have managed to pull out of history's woodworks) has seriously believed in a flat earth since at least the Sixth Century BCE. The other myths that I will discuss in this article are less common, but still present in the minds of many readers: the notion that ancient mariners charted the coastline of Antarctica in the far distant past, and the idea that medieval maps typically marked the edge of the known world with the caption "Here There Be Dragons (or Tygers, or Monsters)." Pointing out what medieval maps are not, therefore, is the primary point of what you are about to read.
Another purpose of this article, however, is to talk about what medieval maps really are. I view this as an essential task, since I cannot convey to you the complexity of medieval cartography without defining the terms involved and showing you a few medieval maps. And I trust that I am not vain if I hope to spur your interest in the Middle Ages along the way.
Between these twin purposes, then, I hope that you find something of interest here and that you walk away knowing something more about maps than you knew beforehand.
A Brief History of Maps
Medieval men and women were neither stupid nor ignorant. If we look at medieval maps expecting the precise geographical accuracy of, for example, a USGS topographical survey, we are bound to come to one of those conclusions. To be blunt, most medieval maps have next to nothing to do with representing geography. This fact places us at several removes from the mindset of a medieval cartographer, since we tend to view maps as little more than geographical representations of the Earth. This expectation is ignorance on our part, not theirs, because maps are never, ever geographically precise.
Jorge Luis Borges is said to have remarked that the only accurate representation of reality would be reality itself; by extension, the only accurate map of the Earth would be the exact shape and size of the Earth itself. Since we cannot construct such a map, we accept a certain level of inaccuracy from our maps, conceding, for instance, that individual buildings need not be marked on a city map, or that fluctuations in elevations will remain unmarked until they deviate by more than a set number of feet from a pre-established elevation. As Borges implied, we must expect some inaccuracies of this kind. But even beyond this simple separation of reality and representation, our society functions in relative naïveté about the accuracy of maps. Most of us blindly accept the borders on our maps, for example, but if you ask an Indian and a Pakistani to draw the boundaries of Kashmir on a blank map of the Indian subcontinent you will almost surely get two very different maps. Closer to home, ask 100 people to outline the Midwest on a blank map of the United States and you will likely end up with 100 different maps. Similarly, most of us are blissfully unaware of the political motivations behind even the prestigious USGS topographical surveys that, while useful for climbing a mountain or finding the elevation of your front door, omit trivial little things like nuclear waste dumps. Maps, like any text or cultural production, carry political motivations, generic conventions, and symbolic associations that must be understood in order to appreciate the map properly. And, as we shall see, medieval maps make this patently clear, as they must be read (and that is the most fitting verb) to be understood.
Most medieval maps were not meant to get someone from one place to another. They were not objects that were carried around in the vest pocket of the weary traveller. Quite to the contrary, most medieval maps were either works of art hanging on cathedral walls meant to show the grandeur of Creation, or illustrations in manuscripts describing the characteristics of Creation. Like most artifacts of the Middle Ages, maps are rare. And, while we know a great deal about medieval maps, any discussion of this kind must begin with the simple caveat that we cannot know about what has not survived. We must work with what we have.
Evelyn Edson, in her book Mapping Time and Space, divides maps into six basic categories, and I will be following her distinctions here. Not only are these categories useful for differentiating the conventions we must understand in reading these maps, but they also serve as a very rough outline of the history of cartography.
Type I: T-O Maps
Fig 1. T-O Map
Full map available at Cartographic Images
Some of the earliest maps that we have are called T-O maps because they look very much like a "T" incised within an "O" (Fig. 1). This "T" shape divides the landmass of the Earth into three continents: Africa, Asia, and Europe. The upright of the "T," separating Africa from Europe, represents the Mediterranean Sea. The right side of the crossbar, separating Africa from Asia, represents the Red Sea. The left side of the crossbar, separating Europe from Asia, represents the Black Sea, the Don River, and the Sea of Azov. The whole map is oriented toward the East so that Asia, twice as large as the other two continents, is at the top of the map. This eastern orientation is common to most medieval maps. The sun rises in the East, and the terrestrial Paradise (the long-lost Garden of Eden) was thought to lie at the farthest eastern point of the globe. Thus, Paradise appears at the very top of most medieval maps.
Though these T-O maps look like representations of a flat, disk-shaped Earth, they are not. The circular shape was merely a convention to represent the concept of the Earth, much as we think conventionally of our planet as a sphere or globe even though it is not. In a similar fashion, the medieval mapmakers intended the circle to represent the top hemisphere of the globe. The bottom hemisphere was generally thought to be entirely covered in water.
As a matter of fact, some medieval thinkers imagined that there were two spheres, one of land and one of water, that God had set slightly off-center from one another at Creation. They imagined the two spheres as pressed against one another, the water soaking into the land. The two spheres did not completely overlap, and one side of the world was therefore dry (with water underground) while the other side of the world was inundated (with land underwater). It was for this reason, they theorized, that no matter how far asea a ship went, it could theoretically anchor into land at the bottom of the ocean.
Other medieval thinkers, however, disagreed with the two-sphere theory, thinking that the two spheres would be unbalanced somehow. God's perfection called for symmetry, and so land at the top of the globe (the T-O world) required land at the bottom of the globe so that the world would be "balanced." This other landmass was called the Antipodes (Latin, meaning "opposite feet") because anyone who visited there would be standing upside down, his or her feet opposite the feet of those on top of the globe. A great many debates went on about whether or not the Antipodes were habitable, and most agreed that they were. But nearly everyone was agreed, however, that even if they were habitable, nobody lived there. More on this when we talk about zonal maps.
Type II: List Maps
Fig 2. List Map
Full map available at Cartographic Images
Somewhat related to the T-O maps are the list maps. These maps take the simple T-O convention and begin to list characteristics of different lands or the different countries and towns to be found in them (Fig. 2). Little attempt is made to place the names or characteristics in distinct geographical relation to one another. Produced today, such a map might represent the United States as a circle inscribed with the names of the fifty states in no particular order.
Type III: Zonal Maps
So why could no one live in the Antipodes? Because of climatic zones. Dating back to classical times, most people thought of the world as divided into zones of climate: a frozen climate at the poles, a torrid climate at the equator, and a blissfully mild (and therefore habitable) climate in between. They logically connected the temperature difference of these zones with proximity to the sun, and to a large extent they were right. Where they were wrong, however, was in their belief that the cold and torrid zones were impassable. It took the voyages of Portuguese sailors down the coast of Africa in the Thirteenth Century to prove, once and for all, that people passing through the torrid zone would not be killed by coming to close to the sun.
Until that time, however, the torrid zone was considered an impassable barrier. The Antipodes, even if they were habitable, could not be inhabited. Like many such concepts in the Middle Ages, the issue was a theological one. Jesus had sent his disciples to preach to all the peoples of the Earth, and if the Antipodes were inhabited and could not be reached then Jesus was giving the disciples an impossible task and was therefore wrong. Jesus, as a member of the Trinity, could not be wrong. So from St. Augustine onward, the simple solution was to view the Antipodes as empty of people.
Fig 3. Zonal Map
Full map available at Cartographic Images
This concept of zones encircling the Earth is represented in zonal maps that try to place locations within the various zones of climate (Fig. 3).
Type IV: Detailed Maps
Fig 4. Detailed Map
Full map available at Cartographic Images
Developing from the simple T-O maps and the information-laden list maps, detailed maps begin to incorporate individual shapes of countries, and geographically-relative labeling (Fig. 4), though any sense of accurate scaling or of complete geographic accuracy is still missing. Symbolic representation remained a key concept in these maps, as the three continents were forced into circular, oval, or even square shapes in order to represent the perfection of God's plan. Scale remained unimportant since maps were rarely used as a means of finding the way from point A to point B.
On the contrary, these maps were used to illustrate what it was like in different parts of the world (climates, countries, populations) and to reveal what might have happened there in the past (the crossing of the Red Sea, the Tower of Babel). Maps often overlaid civilizations, events, and locations from different time periods in order to create, in essence, a history of the world represented in a visual medium. We can see something similar in tourist maps of today: a tourist's map of England might highlight various important historical locations (London, Canterbury, Bath) as well as historical events (Battle of Hastings, Bosworth Field, Runnymede) within the context of today's modern cities, roads, and geography.
Type V: Portolan Charts
Using maps for practical purposes of navigation (getting from here to there) was far from the norm during most of the Middle Ages. It took a burgeoning sea trade and an age of sea explorations to make the next step in medieval mapmaking. Beginning in the Thirteenth Century, then, sea-charts start popping up in various countries of Europe. These charts listed seaways, ports, anchorages, and other information of nautical interest. It appears, for the most part, that prior to the popularity of these so-called portolan (from the Italian portolano, meaning "little port") charts there was no extensive use of such maps.
Fig 5. Portolan Chart
Full map available at the James Ford Bell Library, University of Minnesota
The portolan chart was meant to get a sailor from one port to another by showing coastlines in intricate detail and providing the first aspects of scale in the form of direction and degrees (Fig. 5). Using a compass (which appeared in Europe during the Twelfth Century), a navigator could use his portolan chart to get from one land to another and then find the port by following the coastal outlines on the map. Most of these maps were produced in Spain, Portugal, and Italy: the three countries most active in exploring the ocean during the Late Middle Ages.
Type VI: Scaled Maps
Fig 6. Scaled Map
Full map available at Cartographic Images
From the portolan chart to a truly scaled map, like the maps we are accustomed to seeing today, was a relatively small step. All that was required was a system of scaling that could be used to measure locations on the round Earth. Such a system was found in Claudius Ptolemy's Geography, a Greek text written circa 150 CE that was translated into Latin in the early Fifteenth Century. This text, presenting the concept of latitude and longitude that we all know and love today, provided a bridge between geography and representation that was grounded in systematized theory. Scaled maps became more and more accurate, and systems of mass production like the printing press made maps available to a wider range of individuals (Fig. 6). With increasing frequency, maps came to be used as practical tools for helping someone get from here to there. As we have already seen, however, geographical accuracy is only a relative thing, even today.
Modern Myth I: The Flat Earth
The people who believed in a flat Earth have been few and far between. It does not take a genius to see a ship sail over the horizon or to look down on a plain from a mountain and thereby realize the curvature of the Earth. That Columbus somehow proved that the Earth was round is therefore preposterous. What Columbus did prove (he thought) was that the great hemisphere of ocean between Europe and Asia (remember the T-O maps?) could be crossed. He was right, and wrong. A ship can sail from Europe to Asia, it just has to get around the Americas. Columbus did not understand that he had found new continents; he, along with the many mariners who followed him, thought that they had found a westward route to the riches of China and India in the East.
Our modern myth that medieval people thought the Earth was flat rests in two misconceptions. First, as mentioned above, people have misread medieval maps. In particular, modern readers have seen the T-O maps as representing a flat, disk-shaped Earth rather than the projected hemisphere that they really represent. Second, most modern readers have assumed ignorance on the part of medieval people, just as they have assumed medieval people were dirty and lived in dank and dark hovels. There is little truth in either assumption; both are misconceptions begun in the Nineteenth Century as part of a reaction against the Catholic Church. For a wonderfully concise rebuttal of medieval ignorance on this topic, see Paul Halsall's post on the Mediev-L listserv.
Modern Myth II: Ancient Sea-Kings
Where to start with this one? For those of you who don't know, a man named Charles Hapgood published the book Maps of the Ancient Sea Kings in 1979, in which he theorized that medieval cartographers had very accurate knowledge of the land underlying Antarctica (lands that weren't mapped until modern times). This knowledge, he theorized, could only have been provided by one of two means. The first option would be a long tradition of maps from a long-lost seafaring culture that made accurate surveys of the globe long before the ice caps impeded their mission. The second option would be (wait for it) extraterrestrial visitors who either visited the earth when the polar caps were smaller or whose advance instruments could peer through the mountains of ice. All of this was based on two primary maps, one produced in 1513 by a Turkish Admiral named Piri Reis, the other produced in 1531 by Oronteus Finaeus. Both of these maps show what appears to be the edge of a large landmass across the southern end of the Atlantic Ocean between South America and Africa.
The problems with Hapgood's theory are many, and I shall just list a handful of them here (far more lengthy refutations can be found on the web, one such example being that of Sean Mewhinney). First, most medieval maps contain islands and lands that never existed. The medieval cartographer, when faced with blank space, chose to fill it up with legendary islands or randomly-shaped landmasses. That the Piri Reis and Oronteus Finaeus maps have a large landmass across the southern hemisphere is completely in line with the thinking of the time: such a landmass was required to "balance" the landmasses in the north. Whether or not anyone had seen the landmass made no difference at all: cartographers inserted a squiggly shape just the same.
Second, to match these oddly shaped regions to Antarctica requires a great deal of work. Hapgood, for example, argues that one part of the map (Europe) is provided with one scale and projection, while the Antarctica part of the map is by turns rotated, enlarged, shrunk, and otherwise distorted by a different projection system.
Third, Hapgood severely distorts the accuracy of his tinkered maps. In other words, even if one accepts all the changes that Hapgood made to the maps (changed scale, changed projection, rotating coastlines, etc.), the maps still fail to come close to the accuracy that he claims for them. Put rather simply, Hapgood (and those who have followed in his footsteps) is seeing what he wants to see, not what is actually there.
Modern Myth III: "Here There Be Dragons"
Though the notion that medieval maps marked the edge of the known world with the phrase "Here There Be Dragons (or Tygers, or Monsters)" is wide-spread in our society, I don't know anyone who has ever figured out how the myth was started. To begin with, let me take the time and space here to provide you the complete list of known medieval maps including this rubric in any language: the Lenox Globe (ca. 1503-7), which bears the Latin phrase "hic sunt dracones" (i.e., "here are dragons") on the eastern coast of Asia. That's it. So how the phrase ever gained the legendary status that we assume it to have is absolutely mystifying. Erin C. Blake has compiled a wonderful list of dragon and similar monster references in maps, but even this annotated list fails to uncover any medieval references other than the Lenox Globe. That the source of the myth is unknown, however, does not change the bare fact that it is, nevertheless, a myth.
I hope that the preceding discussion has gone some way toward revealing both the difficulties and the joys of medieval maps, as well as dispelling some of the prevalent myths about them. I might also hope that some of you will allow such differing concepts of cartography to infiltrate your writing. Both fantasy and science fiction are, in part, based on a departure from our present reality. The departure may be simple and subtle or complex and dramatic, but authors of speculative fiction bend the expectations of the reader by creating a new world through which they are given free rein to work their art. Whether these fictional realities are used to comment on our present reality with new ideas, fresh perspectives, and different value systems or merely used to entertain, it is the foundational act of world creation that sets science fiction and fantasy apart from other genres. And the beginning point of world creation (be it fantastic, futuristic, or otherwise) is the determination of the mindset that the inhabitants of that world share, the common worldview that lies behind everything that they do and say. The heart of our fictions lies deep in our own experiences, and it is through examination of our own history that we often mine the material for constructing these divergent worldviews. Knowledge of medieval cartography is only one piece in the puzzle of our own history, but maybe, just maybe, it is the different perspective that will act as a stepping stone for some of you to create new worlds of fiction in which we all can share.
Happy reading and writing.
A native of Colorado lost in upstate New York, Michael Livingston holds degrees in History and in Medieval Studies. His research interests have recently taken him into the field of medieval cartography, though his professional publications include an edition of the Anglo-Saxon poem Wulf and Eadwacer and work on the history of early Christianity. Though he finds time to read and write fiction occasionally, most of his free time is currently spent fly-fishing or walking his dog.
Edson, Evelyn, Mapping Time and Space: How Medieval Mapmakers Viewed Their World, The British Library Studies in Map History, Volume I (London: The British Library, 1997).
Harley, J.B. and David Woodward, eds., The History of Cartography (Chicago: University of Chicago Press, 1987).
Kamal, Youssouf, Monumenta cartographica Africae et Aegypti (Leiden: E.J. Brill, 1935).
Miller, Konrad, Mappaemundi: die altesten Weltkarten, Volume III (Stuttgart: J. Roth, 1898).