Wolrd map by Ranulph Higden, dated 1400

World maps before 1400

How did mappae mundi develop before 1400? P.D.A Harvey investigates.

It is only in maps of the world that we can clearly see a continuous tradition linking Roman and medieval maps. They were the only kind of map that was drawn at all commonly in the middle ages, and well over a thousand survive from the 7th century onwards, far more than all other medieval maps put together. It is not surprising that the word for map in English and in some other languages derives from the phrase used for a world map in the middle ages: mappa mundi. But this did not mean ‘map of the world’ – the idea of a map was only starting to take shape – but ‘cloth of the world': some world maps were painted on cloth as wall-hangings and the phrase came to be more widely applied, first to world maps in general, then to other maps as well.

Indeed, the vast majority of medieval world maps are scarcely maps at all. They are diagrams – diagrams of the world – and are best understood as an open framework where all kinds of information might be placed in the relevant spatial position, not unlike a chronicle or narrative in which information would be arranged chronologically. This information might be of many different kinds. Certainly it included geographical information, and in the 15th century this came increasingly to dominate these mappae mundi, turning them into the sort of world maps we are familiar with today. But in the earlier world maps the geographical element was only one of many: the map was a vehicle for conveying every kind of information – zoological, anthropological, moral, theological, historical.

Moreover, many of these world maps, from all periods of the middle ages, provide just the framework with little if anything added – simple diagrams, often no more than a few centimetres across, drawn as illustrations to philosophical or scientific treatises. Before the 15th century nearly all world maps derive from one or other of two basic diagrams, more or less elaborated. Both show the world as a circle. One has north at the top and is divided by horizontal lines into seven bands, representing frigid, temperate and torrid zones north and south of the central ocean river – the zonal or climatic map. 

13th century zonal world map

13th-century world map

This manuscript map from the Cotton Collection is an example of a 13th-century world map with geographical areas shown in zones.

View images from this item  (1)

The other has east at the top and is divided by a T-shape into the three continents, Asia at the top, Europe on the left and Africa on the right – the T-O or tripartite map. A variant of this is the quadripartite map, which adds a further continent, the Antipodes, on the right, separated by the ocean river from Asia and Africa. Even such large and complicated productions as the 13th-century Ebstorf and Hereford maps may both derive ultimately from the tripartite map.

Diagrammatic world map

Medieval diagrammatic world map.

This manuscript map made in 1075 is a T-O map divided into three continents: Asia, Europe and Africa.

View images from this item  (1)

Usage terms

Public Domain in most countries other than the UK.

Though less elaborate than these 13th-century maps, some much earlier world maps presented far more than the basic diagram – indeed, it would be wrong to see any general progression from the simple to the detailed. The Beatus maps are an interesting early group, so called because they occur in manuscripts of the commentary on the Apocalypse written by the Spanish theologian Beatus of Liebana in the late 8th century. Mostly four-sided, not circular in shape, these Beatus maps illustrate the work of the apostles in spreading the Gospel and particularly prominent is the picture of Adam, Eve and the serpent that marks the earthly paradise.

Silos Apocalypse

12th-century Beatus world map.

This double-page manuscript map from the early 12th century shows the Mediterranean, the Red Sea, Britain and Scotland in different colours and shapes.

View images from this item  (22)

Usage terms Public Domain

All these maps – like most medieval world maps – are built up from the starting-point of one or other of the simple outline world diagrams. The Anglo-Saxon or Cotton map is something quite different. It dates from the early 11th century and was drawn in England; its English origin is betrayed by the name it gives the Bretons, ‘suðbryttas’, spelled with one of the runic letters used in England for th, but is anyway indicated by its location in a volume of Anglo-Saxon writings that belonged in the 17th century to Sir Robert Cotton. It was not, however, compiled in Anglo-Saxon England. Rather it is a copy, perhaps not a very good copy but the clearest we have, of a world map which must have been constructed in the Roman period and which probably contributed much to the geographical elements in many medieval world maps. It may even be that we have here a direct descendant of the world map which Marcus Vipsanius Agrippa, son-in-law of Augustus, compiled at the end of the first century BC and which was probably based on the survey of the world ordered in 44 BC by Julius Caesar. But whatever its origin, the Cotton map is extraordinarily important, and its significance, how far it is related to other medieval maps, has probably not been fully explored. 

Anglo-Saxon Mappa Mundi

Anglo Saxon Mappa Mundi

This map of the world was probably created at Canterbury in about 1025. Like other medieval world maps, it is orientated with east at the top (British Library, Cotton Tiberius B V/1, f. 56v)

View images from this item  (5)

Usage terms Public Domain

That there was such a relationship, that other medieval world maps drew on some version of the Roman map that lies behind the Cotton map, seems fairly certain. Some of its basic shapes – the Mediterranean islands, the strangely elongated Black Sea – occur regularly. We find even its nearly circular Jutland on, for instance, some of the maps in the works of Isidore of Seville and Lambert of Saint-Omer. How far the Cotton map is a faithful copy of its Roman prototype we can only guess. The prototype may well have included the major European rivers, which appear on a number of medieval world maps but only in vestigial form on the Cotton map; and the Cotton map looks as if it may have suffered some distortion in being drawn to fit the particular page. On the other hand, what to our eyes seems one of its more distorted features, the way south-west Britain and north-west Spain are drawn close together, may well be faithful to the original, for some early medieval writers echo the belief that the British Isles were not far from Spain.

Where we might particularly look for the influence of the Cotton map – or, rather, of its prototype – is in the group of world maps that mark the culmination of the early medieval tradition. Dating from the 12th century to the 14th, they have long been recognised as a single interrelated group, though they do not all have the same antecedents and the relationships between them have not been fully worked out. They are very far from uniform. Even in size they differ enormously, ranging from the tiny Psalter map, less than 10 centimetres diameter, to the huge Ebstorf map more than 3.5 metres across. In the larger ones the map was used, more than in any other medieval maps, as a vehicle for every kind of information, learned and moral, and the spatial element becomes little more than the framework of presentation. Many earlier world maps illustrated comprehensive encyclopaedias; these were practically encyclopaedias in themselves.

Map Psalter

Psalter World Map, f9

This 13th-century English map shows Jerusalem at its centre, with a view of monstrous heads surrounding the map.

View images from this item  (4)

Usage terms Public Domain
Held by © British Library

It is in this group of maps that the significance of the phrase mappa mundi, cloth of the world, becomes apparent, for although the survivors are all drawn on parchment we now see the world map being used as a freestanding work of art and learning, hung up on a wall or behind an altar as decoration, as a source of edification and instruction, or as a symbol of deity. One writer who used the phrase in the early 13th century was Gervase of Tilbury; it is significant that he was born in England, for nearly all the maps in this group have English associations. That is not to say they were all produced in England. The Ebstorf map was unquestionably drawn in Germany, but was arguably drawn at the behest of this same Gervase of Tilbury, now living in Germany; or of another Gervase whose origins are unknown but whose name was much more unusual in Germany than in contemporary England. Another map that found its way to the cathedral library at Vercelli in north Italy originated either in England or, perhaps more likely, in the neighbourhood of Paris.

All the other maps in the group are of English origin and it is clear that in the 13th century the large world map, the cloth of the world, was a peculiarly English genre. One of the late 12th century, illustrating the encyclopaedic chronicle of the world by Honorius of Autun, came from the library of Sawley Abbey in Yorkshire and was probably drawn at Durham. Besides those that survive we know of others, now lost, through references in contemporary writings: in the 13th century King Henry III had mappae mundi painted on the walls of his palaces at Westminster and Winchester, while Matthew Paris, monk at St Albans, mentions one that he saw at Waltham Abbey. More may still be discovered. During the last few years fragments of two of these English world maps, hitherto unknown, have come to light, both having been used in the 16th century as bindings for volumes of estate records, one by the college of Bonhommes at Ashridge in Hertfordshire, the other by Walter Aslake, owner of manors in north-west Norfolk. Enough of each survives to identify its date, its English origin and its relation to other maps in the group.

The giant Ebstorf map was destroyed in an air-raid on Hanover in 1943. It has however been redrawn with scrupulous care, using surviving photographs as well as evidence of the colouring of the original so although physical examination of the map – its composition, its writing, its drawing – is no longer possible we have full access to its contents. The Hereford map is now the largest surviving map of the group, but it is much smaller, only 1.3 metres across, so its area is thus only about one-seventh that of the Ebstorf map. They are not very closely related; the Hereford map is nearer to the map from Sawley Abbey, the Ebstorf map to the Vercelli and Psalter maps. Nor are they contemporary. The date as well as the authorship of the Ebstorf map is a matter of debate, but it was arguably a copy made at the end of the 13th century of a map drawn some sixty years earlier; it was made at or for the convent at Ebstorf, near Lüneburg, and came to light there in 1832. The origin of the Hereford map is clearer; it was probably drawn at Hereford in about 1300 on the basis of a map now lost that had been created a little earlier at Lincoln by Richard of Holdingham whom the map names as its author, and apart from short episodes it has been at Hereford ever since. In many obvious ways the maps differ. Thus Ebstorf (but not Hereford) has the head, hands and feet of Christ visible at top, sides and bottom of the map as though the world were his body, while Hereford (but not Ebstorf)  has protruding labels bearing the letters M, O, R, S, death – a reminder of the limits of the material world. Again, to take examples of details, Ebstorf (but not Hereford) has a lifelike camel beside Jerusalem; Hereford (but not Ebstorf) has a charming mermaid in the eastern Mediterranean. But in concept and in the sources they use the two maps are very alike, and they bring together the non-geographical material that was drawn on, in a much more limited way, by the compilers of many lesser medieval world maps.

Foremost among these sources was of course the Bible. Both maps have Jerusalem at the centre – by no means all medieval world maps place it there – but on Ebstorf it is drawn with square walls enclosing a picture of the risen Christ while on Hereford it is circular with Christ crucified outside it. Among places or scenes from the Old Testament entered on both maps are the ark of Noah at Mount Ararat (with dove and olive branch on Ebstorf), Babylon with the tower of Babel beside it, and Mount Sinai beside the Red Sea (coloured red on Hereford, which shows Moses receiving the tablets of stone and marks the route taken by the people of Israel). From the New Testament both show the stable at Bethlehem, but only Ebstorf locates (just a few) incidents in the life of Christ. Both take material from the lives of the apostles and the early history of the church; thus Ebstorf shows the tomb of Bartholomew with a lamp burning above while Hereford shows Augustine standing in the church at Hippo. Much, as we would expect, was drawn from writings about distant lands – descriptive, moral, historical and literary. Among them were Pliny's Natural History, which describes mythical races of monstrous beings, such as the dog-headed cynocephali or the sciopods who used their one enormous foot as a shade from the sun; the Physiologus, a 3rd- or 4th-century compilation which sets out the moral lessons to be learned from the ways of such strange creatures as the phoenix or the pelican; and the medieval romance of Alexander the Great, with its account, embellished with marvels, of his travels to India.

Some of the most interesting details on these maps were drawn from the contemporary world of the compilers. Many towns in Europe are marked and named. It has been shown that places in France on the Hereford map come from a series of medieval itineraries – a route used by Florentine merchants to carry English wool from Bordeaux to north Italy, the pilgrim route to Compostella and other routes from north to south. The Ebstorf map, unlike Hereford, shows a great many towns in Germany, perhaps taken from a series of river-based itineraries. It also has some local detail that seems to come from the compiler’s personal knowledge: at Reichenau, near Constance, where the abbey and two monastic cells appear, and at Ebstorf itself, where we see not only the convent with spire and cross, but also its five martyrs’ graves, the subject of a developing cult in the 13th century. On the other hand the Ebstorf map makes a nonsense of Germany's river system by linking the Weser and the Main; like even worse mistakes on the Hereford map, which seems to join both Weser and Elbe to the Danube, this probably arose from mistaking mountains for rivers when the map was set out. The compiler's own observation of the part of the world he knew really played a negligible part of the world in the construction of either map.

World map by Ranulf Higden

Wolrd map by Ranulph Higden, dated 1400

This map is one of the various versions of a medieval world map in a popular history book compiled by Ranulf Higden.

View images from this item  (1)

Usage terms Public Domain

More or less elaborate diagram-maps in the tradition of the early medieval world maps were still being produced in the 14th and 15th centuries. One of them – in various versions – appears in copies of a compendious and immensely popular history book, the Polychronicon, written in the mid-14th century by Ranulf Higden, monk at Chester. That he was English may be significant given the English associations of the group of world maps that includes the Ebstorf and Hereford maps.

'Liber secretorum fidelium crucis' by Marino Sanudo with maps by Pietro Vesconte

Liber secretorum fidelium cruces by Marino Sanudo. c. 1321.

This map by Vesconte shows a more recognisable outline of the Mediterranean and Black Sea.

View images from this item  (9)

Usage terms Public Domain

Yet if we look at the world maps that Pietro Vesconte was drawing at Venice in the 1320s the scene suddenly changes. The Mediterranean and the Black Sea are no longer an unrecognisable pattern of shapes that can be identified only by the names attached to them; instead we see, more or less accurately drawn, the outline that we are familiar with today. This reflects the advent of the portolan chart, and indeed even the Aslake map, firmly linked as it is with the Ebstorf, Hereford and, especially, the Psalter maps, takes the Canary Islands and the places it names along the African coast from a portolan chart of about 1350. The development of the portolan chart was crucial to the development of the world map in the later middle ages.

Aslake world map

14th-century English world map

The badly damaged Aslake world map shows many interesting African place-names beyond the Nile.

View images from this item  (1)

Usage terms

Public Domain in most countries other than the UK.

This is an updated version of a text first published in P.D.A. Harvey, Medieval Maps, British Library, 1991.

  • P.D.A. Harvey
  • Professor P.D.A. Harvey is a specialist on the economic and social history of Medieval England and the history of cartography.

The text in this article is available under the Creative Commons License.