TITLE:  The Cottoniana or Anglo-Saxon Map

DATE:  ca. 995 A.D.

AUTHOR: from Priscian‘s Periegsis

DESCRIPTION: The Cotton Tiberius is the richly illuminated 11th century manuscript in the Cotton collection of the British Library and contains one of the oldest and most excellent world maps. Called the Cottoniana or Anglo-Saxon map, it dates from 995-1050, just before the Norman Conquest, and does not appear to belong to any one of the identifiable “families” of medieval maps, as described by M.C. Andrews or others.  It is far removed from all of the members of the Beatus group (#207); it is equally far removed from the school to which the Ebstorf, Psalter and Hereford plans appear to belong (#223, #224, #226).  Nor has it any relation with the various types of Zone/Climate maps which are known under the names of Macrobius, Sallust, or Isidore’s T-O diagrams (#201, #205).  The map itself occurs in a copy of Priscian’s Periegesis, a fifth century manual of geography based upon an earlier treatise.  The manuscript that contains the map (Cotton MSS, Tib. B. V.) is made up of various pieces, collected by Sir Robert Cotton in 1598.  The manuscript is bilingual with writings in Latin and Anglo-Saxon, and its English provenance is reinforced by the use of an Anglo-Saxon letter in the term Suo bryttas [Brittany] on the map. The map is on folio 56, and is immediately followed by a copy of Priscian’s Latin version of the Periegesis of Dionysius, De situ terrae Prisciani Grammatici, quem di priscorum dictis exerpsit Ormistarum, written in the same hand as appears on the map.  However, the map stands in no special relationship to the work that it professedly illustrates.  It is, indeed, more closely linked with Paulus Orosius’ Universal History (out of the 146 legends on this map, 75 of which occur in Orosius; 75 contain the textual basis of the whole map and all of its names of countries with few exceptions, see Book IIA, #205). It also has certain obligations to Pomponius Mela’s De Chorographia [Cosmography] of 40 A.D., St. Isidore of Seville, and the topographical writings of St. Jerome. And, finally, it bares some indications of a much later time, the age of the discoveries and the migration of the Northmen in the eighth, ninth, and 10th centuries.  The correspondences of various names and descriptions in Adam of Bremen afford at least a possibility that the former sometimes drew from the same originals as the great northern annalist, while some of the names in the British Isles, in Gaul [France], and in the Far East and northeast, support the 10th century date, which most scholars are inclined to accept.

In its presentation of the world as a whole, this map adopts a roughly square form measuring 21x17 cm, and in this one respect it recalls some of the less desirable aspects of examples of the Beatus Ashburnham derivative Book IIA, #207). However, though of small size, it is one of the most unique of all medieval world-pictures.  In the execution of this right-angled design there is all the difference between the narrow perspective of an uncompromising symmetry found in most of the maps of its period (such as the T-O designs), and a certain respectable, if not highly developed, knowledge and scientific insight is reflected in this map.  It portrays, with comparative fullness and accuracy, various places, regions, and natural features elsewhere omitted, or misunderstood in cartography until a much later date.  In fact there is hardly any map from the Middle Ages, before the appearance of the portolani, which can be compared with the Cottoniana in its detailed delineation of various coastlines. 

The marvels promised by the inscription above the poem are represented on the map by the Cinocephales (dog-heads), the gens Griphorum (a conflation of griffins and people), Gog and Magog, the burning mountain and the mountain of gold, the last two located in remote corners of Asia.

In geographical content, it does follow the medieval European convention of orientation with East at the top and somewhat centered on Jerusalem.  The bulbous projection of land on the coast, north of Jerusalem, is perhaps meant for Carmel. Some idea, though exaggerated, of the Syrtes on the North African coast is evidently possessed by the cartographer.  Its eastern limit is the Persian Gulf and Red Sea.  Asia is shown at right angles to the coast of North Africa, an inferior representation when compared with such earlier, though uncirculated, renderings of Ptolemy.  In Africa, the lakes east and west of the Lacus Salinarum near the north coast are noteworthy; like Brigantia (of Lighthouse fame) in the northwest of the Spanish peninsula.  Mons Clinax [-max] in the middle of the South African coast, is perhaps a misty reference to the “Chariot of the Gods”, as described by Hanno and the Greek and Latin geographers who copied him; while the two small unnamed isles, west of Mount Atlas, are probably intended for the Insulae Fortunatae.

Of purely inland geography, unconnected with the coast, there is not much in the European region of this map: the Huns, Dalmatia, Dardania, Histria, and Tracia, all circling around Pannonia.  What is now European Russia is here contracted to a mere neck of land.  The Caspian Sea, inaccurately opening into the Northern Ocean, is of unusual size. 

However, this is apparently the first map to add to the knowledge of Ptolemy with regards to northwestern Europe.  England, Scotland, Ireland, Denmark (Neronorweci or Neronorroen), and France are better drawn on the Cottoniana than any other medieval map.  The British Isles are prominently depicted with Ireland oriented east-west, vice north-south. Scotland is curiously twisted to the left instead of to the right, as in Ptolemy.  There is also no small comparative merit in the land of the Scrito-Finns and Island, representing present-day Scandinavia, and in Sicily, whose three angles appear; and the north coast of Asia Minor is likewise good.  On the other hand, the western Mediterranean is rather crude and very contracted.  France is so squeezed between Spain and Italy that its south coast almost disappears, except for the Gulf of Lyons, which is fairly well delineated.  In

Greece the name Macedonia seems to be written over Morea; Athens and Attica are widely separated.

It is sometimes claimed that the association of Armenia with Mount Ararat and Noah’s Ark appeared on maps only after the First Crusade. In this map, which was made in ca. 995, some 100 years before the First Crusade, Mont Ararat and Noah’s Ark are shown, firmly placed in the territory of Armenia.

In the map mountains are shown green; red is used for the Persian Gulf and the Red Sea, as well as the Nile and some other rivers. At the top left, behind the Lion we see the Taurus Montes, where the Tigris and Euphrates have their sources. Below the range is the legend Montes Armenia, which describes the twin Peaks of Mount Ararat; here shown sideways, with the three-storey Arca Noe [Noah’s Ark] perched on top. Below the Ark the legend Armenia can be seen, although somewhat masked by the print bleeding through. These shapes are reminiscent of Orosian maps. Further down we see a second range of mountains also named Taurus, the continuation of the remaining section of the same range, continued with a break. Armenia’s neighboring provinces are Hyberia and Mesopotamia. To the left of the Ark, the intricately shaped bay with two islands is the Caspian Sea. Nearby is the region of containment of the tribes of Gog and Magog, situated near the northern ocean.

Inside the area between the Black and Caspian Seas are two legends. One reads Mons Albanorum [Albanian Mountains, possibly the Caucasus] and the other is regio Colchorum, the region of Colchis, located northeast of the Black Sea. Hiberia [Iberia] is shown south of Montes Armenie, between the two rivers rising form the Armenian plateau and Taurus, in the territory of Mesopotamia.

This is one of the few medieval maps that shows divisions of provinces and countries, indicated by straight lines, though the general T-O shape is still preserved by the dominant body of the Mediterranean, here filled with a multitude of islands, complemented by the Nilus in Africa and Tanais [Don] at the center left of the map, with its source in the green mountain. The Nile is shown in two sections. Flowing eastward from Upper Egypt, it turns west and somehow disappears underground to emerge further down and continue its flow towards Alexandria and the Mediterranean Sea.

In Asia there is much more inland geography, chiefly connected with the Twelve Tribes and Biblical history. To the west of the Caspian Sea can be found Gog and Magog, adjoined by the Turchi; the Bulgari is placed between the Danube and the Arctic Ocean; and Taprobane [Sri Lanka] occupies the place usually given to the Terrestrial Paradise. The Cottonian map places Gog and Magog hard by the northern ocean, west of the Caspian Sea and the Ten Lost Tribes appear in the Middle East. The map’s author had not yet identified these peoples with one another. The two legends are treated separately, as in patristic literature.

A lost Roman province map may have been the source of the divisions so clearly marked in Asia Minor, in Central and Southeastern Europe, and in North Africa.  The Biblical loans may be traced in many names and also in certain aspects of the general plan.  Indeed, it is obvious that here the design was not merely indebted to the Scriptures for details that almost all medieval maps exhibit, but it was also, to a large extent, devised for a special Biblical lesson - a picture of the settlement of the Twelve Tribes of Israel.  Other Biblical connections, traceable especially in the center of the map, include (directly or indirectly) not only the nine of the twelve tribes of Israel and Jerusalem, but also Bethlehem, Babylon, Tarsus, Caesarea Philippi, and the Ark of Noah (among the vignettes), the Hebrews’ passage through the Red Sea, the lands of the Philistines, Amonites and Moabites, Mount Pisgah (here Fasga), and Galilee.  Paradise, uncharacteristically, is not shown. Most of the Biblical names found on the Hereford, Lambert, Henry of Mainz, the Psalter, and Ebstorf maps are perhaps, in many cases, borrowed directly from this earlier Anglo-Saxon work. 

There are several names and features which show striking independence of any other known map authority of the earlier Middle Ages. Among these are five names in Britain: Camri or Cambria, and Marinus-portus in the northwest; Kent, London, and Winchester on the southern shore; and Arma or Armagh in Ireland; the Sud-Bryttas [South Bretons] in northern Gaul; the Golden Mountain of the Far East and the Boreani and abundant lions of the northeast of Asia. Major cities are represented by drawings of fortifications (London and Armagh (Ireland)). In addition, there are a number of unlabelled provinces, rivers and islands, leading one to surmise that this map was copied from a larger and more detailed map.

The comparative excellence of the Cottoniana is perhaps due to its being the production of an Irish scholar-monk living in the household of the learned and traveled Archbishop Sigeric of Canterbury (992-994), with whose Itinerary, from Rome to the English Channel, the present design has several curious resemblances.  In the British isle of the pre-Norman period, there is no school of learning art, or science comparable to that which sprang from the Irish Church of Patrick, Colomba, and Ardan; and the insertion of the name of Armagh, so rarely found in medieval maps, strengthens the view that there we have the handiwork of a student who was trained in Irish schools, or derived his knowledge from men so trained.

The coloring of the Cottoniana is grey for most seas; red is used not only for the Red Sea (top right), but also for the Persian and Arabian Gulfs, the Nile with its delta and other African rivers and lakes; and bright green is used for all mountains.  The handwriting is small and difficult, with peculiar formation of various letters, i.e., C is written like an R, O like A, R like P and A.

Considering the 11th century Anglo-Saxon mappamundi, the oldest such English map surviving, as a form of a virtual world more analogous to a digital environment than physical geography can reveal much about this famous map’s cultural mechanics and meaning. As not merely a measurement, but as a creation of the world known to Anglo-Saxon England, this mappamundi charts a number of struggles between the two realities of England’s marginal locus in the historical record of the known physical world, and of the Anglo-Saxon impulse to re-center the world on their own island.

In past discussions, the cultures of medieval worlds such as Anglo-Saxon England have been understood to lack the veneer of ideological unity that comes with later examples of more modern and overtly ratiocinated expressions of national identity. Ernest Gellner, for instance, uses explicitly spatial figures to describe such conflicting realities; for Gellner, ‘pre-modern and pre-rational’ visions of the world lack ‘a single continuous logical space’, and instead consist of ‘multiple, not properly united, but hierarchically related sub-worlds’, and ‘special privileged facts, sacralized and exempt from ordinary treatment’.

The Cotton Map certainly appears to fulfill Gellner’s formula: it contains, on the whole, a rather fractured and jumbled version of the known world, crossed by lines and marked by inscriptions drawn from a skein of classical, Christian, legendary and local traditions of cartography, and, on the face of it, refuses ‘a single continuous, logical space’. The map likely derives from classical maps and/or a version of Orosius’ in Historiae adversum Paganos; unsurprisingly, the Anglo-Saxon worldview also includes the hierarchical nature of the sub-worlds and sacred places it contains. Rome and Jerusalem appear prominently – with its six towers Rome is one of the two largest cities shown (along with the historically massive Babylon), biblical lands occupy the center band of the map, marked by straight, confident lines, holy waterways are distinctively colored in bright red, and a paradisiacally described island occupies the top centre of the map. England, in the meantime, occupies the lower left corner of the map, its cities tiny and its terrain free of the distinctive straight lines that dominate the center strip of the Holy Land.

However, the Cottoniana map also encodes a certain cultural unease with this represented and sacral order, a discomfort revealed by how the map’s virtual nature allows what is edge and what is center to fluctuate over its surface, and how this fluctuation articulates an Anglo-Saxon challenge to its homeland’s traditional place in the geographical order of the world. Anglo-Saxon writers inherited a long historical tradition of Britain’s peripheral status as a remote corner of the world. In the first century, the Roman geographer Pomponius Mela described the inhabitants of the island as inculti omnes and ita magnis aliarum opum ignari [‘all uncivilized’ and ‘moreover ignorant of a great many other things’]. Mela correlates the ignorance of Britain to the island’s extreme distance from Rome and the Continent, and rounds out his description by noting that the British pecore ac finibus dites [‘are rich only in cattle and land’].

In addition, Mela’s use of finibus for ‘land’ may also contain a further joke at the expense of Britain’s remoteness; finis commonly carries connotations of limits, ends or borders. The people of Britain, therefore, were rich only in livestock and their own place on the margins.

Three hundred years later, the Roman historian Solinus repeats this sentiment, noting that for all practical purposes, the coastline of Gaul stood as the edge of the known world, while Britain represented a land beyond the periphery – paene orbis alterius [‘almost an other world’]. Such views continue through early medieval writers such as Isidore, and early native writers remained substantially invested in such views of their home.

In the sixth century, Gildas writes, Brittannia insula in extremo ferme orbis limite circium occidentemque [‘the island of Britain lies practically at the extreme limit of the world, towards the west and the north-west’]; by the eighth century Bede has softened, but not erased this perception: Brittania Oceani insula, cui quondam Albion nomen fuit, inter septentrionem et occidentem locata est, Germaniae Galliae Hispaniae, maximis Europae partibus, multo interuallo aduersa [Britain, once called Albion, is an island of the ocean and lies to the north-west, being opposite Germany, Gaul, and Spain, which form the greater part of Europe, though at a considerable distance from them].

As Nicholas Howe has recently argued, native writers’ positioning Brittania in the north-west demonstrates the historical influence of the Roman vantage point, but, in the shift from imperial Rome to medieval Rome, this distantiated perspective assumes Christian as well as geographical significance. In Gildas, one finds a striking example of spiritual calibration, as the historian initially refers to the island as divina statera [in the divine scales] and then describes the conversion of Brittania as the sun’s warming of glaciali frigore rigenti insulae et velut longiore terrarum secessu soli visibili non proximae [an island frozen with lifeless ice and quite remote from the visible sun – recessed from the world]. If, as Orosius describes in Historiae adversum Paganos, world history did move from East to West, then readers of his Anglo-Saxon translation could easily have found themselves on the very edge of one of feower endum pyses midangeardes – at the edge of the physical world and at the end of the processes of history and salvation. The Orosian ‘four ends of this earth’ also recall the common scriptural phrase ‘the four corners of the earth’, a phrase translated directly (feowerum foldan sceatum) or alluded to many times in Old English literature.

The Cotton Map makes a sure statement of the Anglo-Saxon geo-historical paradox: a major part of Anglo-Saxon identity is particularly rooted in Christian authority, whether shaped by fears of pagan conquest or a desire to prove ethnic superiority by converting others, but this ideology historically has in turn viewed England, literally, as the edge of nowhere. At least on the surface, English cosmography does not, in the words of one commentator, ‘develop strategies for recuperating auctoritas’ until well into the 13th century. In its depiction of England, however, the Cottoniana map anticipates later strategies for transforming the cosmographical auctoritas of the Anglo-Saxon homeland, and pushed back against the traditions that have relegated it to the edge. The edge of geographic knowledge, Mary Campbell reminds us, can be a location charged with ‘moral significance’ and even ‘divine dangerousness’. Living on the edge, as it were, Anglo-Saxons felt the danger of their borders keenly throughout their history. The time of the Cottoniana map was no different, as the Anglo-Saxon state began the 11th century unsuccessfully fighting off one set of invasions, and did not last the century fighting off a second set; but from at least the time of King Alfred, as the Anglo-Saxon state was formed, it also coalesced around a cultural center other than Rome or Jerusalem. And even as the Cotton map acknowledges the liminal world that is the origin of the Angelcynn, it also promotes their geographic desires and anxieties that England, ultimately, should also be a center, not a corner, of the world.

J. B. Harley has observed the widespread practice that ‘societies place their own territories at the center of their cosmographies or world maps’, but a quick glance at the Cotton map confirms that it does not, on the page, center Anglo-Saxon England.  Such a failure derives in part from sheer spatial necessity; as far as Anglo-Saxons were concerned, precious little world existed west of the island, making the centering of England impossible in terms of simple layout. Yet the map, itself a product of late Anglo-Saxon culture, appears to practically embrace the traditional marginalization of England. While Anglo-Saxon England possessed a coherent national identity since at least the ninth century, this 11th century depiction refers to the island by its Roman name, Brittania, rather than the more contemporary Angelcynn. In the ‘big picture’ of the mappamundi, England appears literally as one of the feowerum foldan sceatum, tucked far away from the center of the map, with only Ireland and Tylen, or ultimate Thule, closer to the physical corner of the manuscript page.

From a literary, if not literal standpoint, then, the Cottoniana map remains centered on its inheritance of Roman geography. However, it is not uncommon for critics to assume that, as a mappamundi, the map also theo-graphically depicts Jerusalem as its physical center. Jerusalem has a long scriptural and exegetic tradition (including Anglo-Saxon writers such as Bede) as the exact center of the earth, and appears as the precise center of a number of famous mappaemundi, most notably the Hereford (#226), Ebstorf (#224), Psalter (#223) and Higden (#232) maps. Importantly, though, the Cotton map should be considered as centered on Jerusalem in only the loosest sense of the term, since the map locates the city slightly down (west) and far to the right of center. Recent commentators have noted that most early medieval world maps actually do not center Jerusalem, and that this convention probably derives from the later political context of the Crusades’ quests for the Holy City.  It is tempting, however, to also consider the de-centering of Jerusalem in the Cottoniana map as a function of the map’s own textual convention, and the way it responds to and ultimately rejects notions of England’s marginalized state.

Again, one of the most striking features of the Cottoniana map is that, unlike most other mappaemundi, its shape is rectangular rather than round. Church writers long agreed that the earth was spherical, but also had a tough job reconciling the circular form of the earth and the scriptural concept of the world’s four corners. In mappaemundi, the usual solution consisted of placing a spherical map inside a square, and filling these ‘corners’ with iconographically suitable adornment. The Cottoniana map, in contrast, has the look of a round map that has been deliberately stretched to fit the dimensions of the manuscript page. The map, while retaining some suggestive quality of roundness, definitely has corners, in marked contrast to the traditional circular format of such maps, including later and more elaborate English mappaemundi, where the British Isles end up compressed, squashed, really, into the curvature of the border.

Fitting the world to the world of the page also allows the mapmaker more space in the lower left-hand corner to depict ‘the angle of England’; as Howe has demonstrated, the doubled meaning of ‘angle’ and ‘Angle’ was not unknown in contemporary descriptions of Anglo-Saxon England. Cornering the world, though, also adds space that the Anglo-Saxon artist did not know quite how to accommodate. Consequently, Eastern Europe and Asia Minor expand considerably, creating vast empty spaces that in turn push Constantinople further north than it is normally found on medieval maps and, likewise, Jerusalem further south. In comparison to later mappaemundi, such a choice actually de-emphasizes the center as it creates room in the corners. In the Cottoniana map, the physical center of the world appears rather sparse, consisting of mostly empty sections of lower Syria and the eastern Mediterranean. Each of the corners, by contrast, is distinguished and distinct: the separate landmass of the British Isles in the northwest, the giant drawing of the lion in the northeast, and the fiery waters and mountains of the southern corners.

Not surprisingly, the British Isles are rather well represented in the map – the one cartographic detail that has occasioned critical comment in the past. The coastline and shape of the islands are pretty particular for the time; they contain icons of three named cities (London, Winchester and Armagh), and a number of delineated regions, including named regions for Ireland (Hibernia) and Scotland (Camri). The Orkney Islands also make an appearance, as, it seems likely, does the Isle of Man. One city in the southwest lacks an inscription, perhaps an indication of the provenance of the map or mapmaker.

Less discussed, however, is the map’s use of water to relate England to the rest of the world. In keeping with classical convention, the map circumscribes the earth with a grey wash of ocean that in effect presents a third frame inside the double-lined border and then the manuscript page. Unlike any other mappaemundi, though, water similarly frames the British Isles, in effect creating the same representation of the world in microcosm. In the formal elements of the map, the curve of the Channel and North Sea echo the form of Mediterranean and Black Seas, which clearly demarcate Europe from the rest of the world. Looking at the map in this fashion, one can then see a succession of three nested identical L-shaped frames, which draw the eye of the viewer from the whole world, to Europe, and then finally, to England. In this scheme, Britain itself then maps its own marginality on to Ireland, which ends up inside a fourth L-shaped frame of water northwest of Britain – a move that to a degree displaces Brittania as one of feowerum foldan sceatum, and brings England further in from the edge.

Since the center of the Cottoniana map’s world presents the general impression of empty space, rather than of a loci medii, the center of Europe likewise appears largely vacant, prompting Patrick McGurk to hazard that the confused jumble of regions and tribes thrown into Central Europe represents the attempt to break up ‘the largest blank area in the map’. In Europe, notably, major cities appear only on its periphery: Constantinople in the east, and Rome and a number of other Italian cities in the south. In sharp contrast, the broad strip of Western Europe that borders England is remarkably devoid of any inscription or detail. Unlike the mappaemundi closest in time to it, the Cottoniana map completely ignores lower Germany and France, and from what appears to be Jutland to the Pyrenees mountains, the map includes exactly one inscription: sudbryttas (discussed below). Of course, due to a greater degree of familiarity with the contemporary state of Frankish and Gaulish regions, the Anglo-Saxon mapmaker may have simply omitted the Orosian and/or source map description of this region as outdated, though it is puzzling that the classical geography was not then replaced with something else. But the blankness of this area can provide two functions: first, it reinforces the map’s accentuation of England by providing another curving blank space, this time of land, to frame the British Isles. Second, the map’s omission of England’s closest continental neighbors might not only reveal an eagerness to distance Anglo-Saxon England from its classical definitions, but also perhaps consciously deny a particular aspect of its current political situation.

In classical and early medieval definitions of Britannia, the territory was long defined as a cosmographic and cultural “other” to continental lands. In his Etymologies, Isidore of Seville, in language closely matching Orosius, explains that Britain is an island within sight of Spain, but literally opposite Gaul: Brittania Oceani insula . . . haec adversa Galliarum parte ad prospectum Hispaniae sita est. Bede echoes this sentiment, noting that in relation to Spain, Gaul and Germany, or ‘greater Europe’ (maximis Europae) Britain stands ‘at a great distance against them’ (multo interuallo aduersa). From Isidore and Orosius, Bede specifically preserves the term adversa, the word fraught not only with locative, but potentially negative connotations (e.g. refutation, misfortune, hostility or punishment) as well. Gallia commonly occurs in most other medieval maps of this region, both earlier and later than the Cottoniana map, precisely because of this marginalizing textual tradition. The Cottoniana map, however, largely resists such characterization as adversa maximis Europae, choosing both to elide the classical inscriptions of Gaul and Germany, and then to leave the region almost entirely void.

What remains most startling about the map’s treatment of Europe, though, is the sole inscription allowed to border England, sudbryttas, presumably meant to represent Brittany. The very form of the inscription reveals much about the Anglo-Saxon attitudes behind it; sudbryttas contains a unique use of the Anglo-Saxon “ó”, one of the only distinctively Old English characters in the text of the map. The literal meaning of the inscription, ‘south Britain’, assumes a somewhat colonialist attitude towards Brittany, and onomastically centers the perspective of the region squarely on England. Such an attitude also references one of the first major cultural events of post-Roman Britain, namely the victory of Anglo-Saxon invaders over native Britain, and the subsequent late 5th century settlement of Bretons in southern Gaul. Thus with reference to the area of classical Gaul, or contemporary France, Normandy, Flanders, Maine and Burgundy, the map chooses to depict a period both after the fall of Rome, and before the rise of Western European political states that would by the middle of the 11th century definitively end Anglo-Saxon power. In this sense, the mappamundi eerily refuses to recognize the very regions that will directly enable the Norman Conquest of Anglo- Saxon England, only decades (perhaps less) after the map was made.

Rejecting classical Otherness, the map denies the geography of the pre- and early Anglo-Saxon literary history that was invested with what Homi Bhabha has termed ‘colonial mimicry’ – the desire of the colonized to present itself as ‘a reformed, recognizable other’. Instead, the version of the Anglo-Saxon world celebrates the origin of Anglo-Saxon culture, and in turn highlights northern, not southern continental connections. In contrast to its treatment of Western Europe, the Cottoniana map lavishes attention on the Scandinavian north, and provides seven names for the area (mostly absent from Orosius), including names for tribes in Norway, Finland, and possibly Iceland. The one inscription from Orosius retained in this area, Daria (Dacia) ubi et Gothia, only intensifies the Cottoniana map’s regard for the Nordic. McGurk notes that the map pulls Dacia considerably out of position; Orosius places Dacia et Gothia in the middle of Eastern Europe, between the territories of Alania and Germania, while the Cottoniana map moves Dacia et Gothia much further north. As Stephen J. Harris has recently demonstrated, Anglo-Saxons likely viewed the Daci as synonymous with Dani (the Danes), and Gothi (derived from Getæ) as the same as the Geats, the famous, if somewhat a historical Scandinavian tribe of Beowulf fame. Likewise, Scithia, here positioned east and slightly north of the island containing the scridefinnas (who are also mentioned in the Old English Orosius), might also have a presumed Scandinavian affiliation.

The Scandinavian elements of the Cottoniana map are not surprising, of course, given the sustained presence and development of Anglo-Scandinavian culture in the second half of Anglo-Saxon England’s history. Accordingly, these elements compare to substantially older examples of Anglo-Saxon literature. The cumulative effect of all these Scandinavian – or reputedly Scandinavian – references recall the famous interpolation of the voyages of Ohthere and Wulfstan in the Alfredian Old English translation of Orosius. Traditionally, the interpolation of allegedly firsthand accounts of explorations of Scandinavia and the Baltic region into Orosius’ classical geography is viewed as a natural extension of ninth century England’s connection and interest in things Scandinavian. Sealy Gilles, for instance, claims that in the Alfredian additions, the Anglo-Saxon Wulfstan’s account of pagan customs persisting on the margins of European Christendom is reminiscent of the ancient customs of the English themselves.  As such, his exploration of a Baltic people authenticates his own heritage and blunts the thrust of Orosius’s Christian apologetic, explicitly written adversum paganos.

Such classical revisionism is an arguable reading, of course, but like the Cottoniana map, the interpolations in the Old English Orosius also present a cultural perspective that centers Anglo-Saxon England, and offers up other regions to be marginalized in its place.

In addition, though these pagan insertions do qualify the original intent of Orosius, Stephen Harris has argued that other Old English alterations to Orosius’ history do not refute Christianity, but rather present ‘a sense of Germanic community [that] shapes the Latin into an Old English story of the origins of Christendom’. Harris’ chief example of this reshaping, namely how the Old English version substantively alters the Goth sacking of Rome, holds particular interest with regard to the discussion of the Cottoniana map. The Old English rewriting of the sack of Rome carefully excises from the original text any emphasis on the political power and spiritual centrality of Rome. Instead, the Anglo-Saxon version uses the occasion of the sacking of Rome to manifest a distinctly Germanic notion of Christendom and kingship, one which ‘permits vernacular access to and Anglo-Saxon identification with an order of identity that has left the senate and people of Rome far behind’.

The Old English Orosius, in effect, de-centers Rome from a distinctly Roman history of the world. In a similar manner, the Cottoniana map shifts the focus away from two traditional centers of the spiritual and geographic worlds on whose margins Anglo-Saxon England existed. We have seen how the physical world of the map literally de-centers Jerusalem – a position it will not manifest for centuries in English mappaemundi – while likewise marginalizing the effect of Rome upon England. One cannot, though, deny Rome a certain presence in the Cottoniana map. Italy alone contains seven cities, including Rome – more than any other region on the map – and the end result is a mass of iconographic power pointing at this imperial and spiritual seat. At the same time, however, the Cottoniana map enacts a strategy similar to that found in the Old English Orosius, and emphasizes the Anglo-Saxon world even as it works within a distinctly Roman source. As noted above, the L-shaped England echoes the forms of the larger geographies that contain it, that of Europe, and then that of the world. Likewise, in England, the cities of Winchester and London, the twin capitals of the Anglo-Saxon world, hold the same relative position as does Rome and its lesser cities in the shape of Europe. Rome may still dominate the cartography of Europe, but, in the framed world of England, London and Winchester occupy the same space, and suggest a willingness and desire to assume a similar role. As Harris points out, King Alfred’s ninth century political victories over Danish enemies in England validate Anglo-Saxon Christendom over pagan beliefs from another edge of the world. Similarly, Nicholas Howe chronicles how Anglo-Saxon missionaries journeying ‘back’ to Germanic heathendom replicate the pattern of and then supersede Augustine’s original mission from Rome to Canterbury. In both examples, Anglo-Saxon activities work to define new cultural edges in relation to an understood English centre. The Cottoniana map operates as a graphic analogue to such literary and conversionary efforts.

In the 11th century ‘real time’ of the Cottoniana map, the imperial glory of Rome is no more real than that of Babylon. As a virtual world, its temporal aspects are no more rooted in the primary world than are its spatial aspects. Rather, the Cotton Map presents several versions of Rome, and of England, at once. For Anglo-Saxon England, Rome meant many things, and the map should be understood as embodying all of them: the past imperial power, responsible for the historical view of England, as well as a present fallen one, and the present spiritual center of Christian belief. Similarly, the map produces many Englands: the past, Othered colony of Roman conquest and then missionaries, the present geographic island, and, most importantly, the desired stable political entity in the process of moving in from the edge of the world and assuming a centric role in, at the very least, a larger corner of Europe.

The Cottoniana map has not been dated more surely than the first half of the 11th century, and possibly may be even slightly later; see McGurk and Dumville, pp. 33–4; on paleographic evidence, the editors date the Cotton Tiberius B. v manuscript to the second quarter of the 11th century, closer to 1050. However, the editors also consider, tentatively, the possibility that some of the script of the map is of a slightly later date, and possibly as late as the early 12th century. Conversely, McGurk and Dumville also consider that this script may also be the work of the manuscript’s main scribe (p. 30). N. P. Ker dates the manuscript to the first quarter of the 11th century, though he notes that the composition of some of the material (e.g. an episcopal list and Sigeric’s itinerary) dates from ca. 990. See Catalogue of Manuscripts Containing Anglo-Saxon (Oxford: Clarendon Press, 1957), item 193, pp. 255–6. The elision of contemporary France may possibly derive from a combination of a Roman source and an understandable emphasis on Scandinavian regions. If the map was drawn around 1050, this elision may also express the negative attitudes towards Normandy developing rapidly among some Anglo-Saxon factions. Even if the absence of France and Normandy is a product of early ninth or 10th century Anglo-Saxon cultural concerns, this omission certainly could have taken on new meaning around 1050.

The 'Anglo-Saxon world map' contains the earliest known, relatively realistic depiction of the British Isles. It was created, probably at Canterbury, between 1025 and 1050 but is probably ultimately based on a model dating from Roman times. This showed the provinces of the Roman Empire, of which Britannia [England] was one. The map was revised and updated in about 800 and again in about 1000. New information was added but at each stage errors and misunderstandings occurred in the copying process.

Like most early maps, this one has East at the top. Nevertheless the British Isles (bottom left) are immediately recognizable and the Orkneys, the Scillies, the Channel Islands and the isles of Man and of Wight are shown. The tortuous shape of Scotland is particularly well drawn. London, the Saxon capital of Winchester and Dublin are indicated using Roman-style town symbols. The size of the Cornish peninsula is exaggerated, probably reflecting the importance of its copper and tin mines in the ancient world. Most tantalizing of all is what appears to be two fighting figures in the peninsula. Could they refer to the conflict between the Saxons and the native Britons in the centuries following the departure of the Romans early in the fifth century, which gave rise to the legend of King Arthur?

LOCATION:   British Library, Cotton MS. Tiberius B.V., fol. 56v.

SIZE: 17.6 x 21.2 cm


*Beazley, C., The Dawn of Modern Geography, volume III, pp. 559-563, 608-611.

*Edson, E., Mapping Time and Space, How medieval Mapmakers viewed their World, pp. 74-84

Edson, E., “The oldest world maps: classical sources of three VIIIth century mappaemundi”, The Ancient World, 24, 1993.

Foys, M., “The Virtual Reality of the Anglo-Saxon Mappamundi”, Literature Compass 1, 2003.

*Harvey, P. D. A., Medieval Maps, Plate 19.

Miller, K., Die Altesten Weltkarten, Heft III (Stuttgart: 1895).

*Galichian, R., Countries South of the Caucasus in Medieval Maps: Armenia, Georgia and Azerbaijan, p. 133.

*Harwood, J., To the ends of the Earth, 100 Maps that changed the world, pp. 35-36.

*McGurk, P., Dumville, D. N., Godden, M. R. and Knock, A. (eds), An Eleventh-Century Anglo-Saxon Illustrated Miscellany: British Library Cotton Tiberius B.V, Pt 1. Early English Manuscripts in Facsimile 21 (Copenhagen: Rosenkilde & Bagger, 1997).

*Nebenzahl, K., Maps of the Holy Land, p. 30, Plate 8.

*Tooley, R. V., Maps and Mapmakers, pp. 13, 48.

* Glorious Age of Exploration, p. 168.



For the most recent comprehensive survey of the mappamundi, see P. McGurk, D. N. Dumville, M. R. Godden and A. Knock (eds), An Eleventh-Century Anglo-Saxon Illustrated Miscellany: British Library Cotton Tiberius B.V, Pt 1. Early English Manuscripts in Facsimile 21 (Copenhagen: Rosenkilde & Bagger, 1997).

See K. Miller, Die Altesten Weltkarten, Heft III (Stuttgart: Jos. Roth’sche, 1895), p. 35 for a discussion of sources and analogues of the map’s inscriptions. Over half of the 146 inscriptions may be found in Orosius.

E. Edson, “The oldest world maps: classical sources of three VIIIth century mappaemundi”, The Ancient World, 24, 1993, pp. 169–84, at p. 182.

The Cottoniana or Anglo-Saxon Map, c. 995, 21x17cm

Detail of Western Europe and Africa