Balkanalysis on Twitter

Exploring Byzantine Cartographies: Ancient Science, Christian Cosmology, and Geopolitics in Byzantine Imperial Mapping

June 14, 2009

By Alex G. Papadopoulos, Ph.D.

Department of Geography

DePaul University, Chicago

This short paper on Byzantine maps and geographic science was born out of a conversation with Christopher Deliso, director of We agreed that there is a need to look at southeastern Europe – the Balkans – from a spatial (geographical) analytical perspective. Our understanding of the region’s historic and contemporary nationalisms, its ethnoreligious and ethnolinguistic identities, and the statecraft that defines the region’s geopolitical peculiarities, can be improved by studies of the production of cultural and political spaces (territorial states, homelands, heritage spaces, symbolic landscapes, and pre-modern Empires, to name a few). I would like to start that conversation about Balkan spaces by looking at Byzantine cartography and geographic theory as political, cultural, and scientific products that shaped and continue to shape our perception of the region.

I suggest that the manner in which Byzantines and others conceptualized geographical space in both symbolic and practical terms has much to do with the way territories were scripted and integrated into cosmological, Earth-based, regional, and political worldviews.

More importantly for us today, geographic and cartographic thinking of, and artifacts that date from, antiquity to the modern era have at some times informed about, and at other times attempted to obscure, the ground truth in the service of politics, power, and ideology. My tasks here are to untangle and describe to the reader a set of geographic and cartographic traditions that, although situated in the remote past, constitute root sources for the way the Byzantines looked at the World, generally, and their region in particular.

The question of chronological provenance

Some clarification about terms: A traditional approach to classifying the maps and their provenance chronologically as “Eastern Roman” and subsequently “Byzantine” would be to adopt as a time frame the establishment of Constantine the Great’s reign (324CE) to the fall of Constantinople (1453CE). There are some pitfalls, however, to adopting a rigid chronological frame, and situating these artifacts and theories firmly as byproducts of the societies and polities of those centuries.

To put it simply, one of the most important lessons of studying Byzantine cartography as science, and Byzantine maps, as material objects, is the diachronic character and influence of the source material on which they were based:

For example, Classical, Hellenistic, and imperial Roman geographic knowledge and science need to be understood as a continuous scientific tradition, expressed through the thought and works of Anaximander (610-546 BCE) (figure 1a), Hecateus of Miletus (c. 550-476 BCE), Herodotus (c. 484-425 BCE), Eratosphenes of Cyrene (276-195 BCE), Strabo (c. 63 BCE-24 CE), and Claudius Ptolemy (c. 90-168 CE) to name some of the most important.

Critically important for us, cartographic works and theories of this long-standing scholarly tradition contributed greatly to geographic and cartographic concepts of the Middle Ages in Europe’s West, in the Roman and Byzantine East, and in the Arab World after the 7th century CE.

Standing on the shoulders of pre-Socratic and Classical scholars, Claudius Ptolemy’s work can easily be counted as the most influential. Grounded far more on empiricism than religious symbolism, his Geography dominated regional cartography and scores of derivative portolan (navigation) cartography into the 16th c. Known to Arab scholars and cartographers, the Ptolemaic opus reaches Florence ca. 1400 and revolutionizes cartography in the West (figure 2).

At the other end of the chronological spectrum, the Fall of Constantinople in 1453 may be construed sensu stricto as the end of Byzantine scientific study of geography, and the culmination of Byzantine cartographic production. Yet, it would be an error not to consider and assess the influence of Byzantine emigre scholars, like Gemistus Pletho who lectured in the 1430s in Florence before returning to Mystra in the Peloponessus, and Markos Mousouros and Zacharias Kallergis, who taught in Rome’s Quirinal College in the early 1500s, where publisher and cartographer Nikolaos Sophianos was a student.

These men were representatives of a learned elite with strong intellectual, and at times, political links to the fallen Byzantine State. Many of them settled in centers of learning in Italy-Rome, Venice, and Florence-in the decades after the Council of Ferrara-Florence (1438-39) attempted the union of the Western and Eastern Churches into an Oecumenical church.

As far as the transmission of geographical knowledge from Byzantium to Italy is concerned, there are significant examples: For example, the scholar Gemistus Pletho was significantly responsible for re-introducing Strabo to the West through his scientific treatise of Strabo’s Geography. The treatise is preserved in Codex Marcianus graec. 379, in St. Mark’s Library, Venice, which most likely belonged to Pletho’s famous student Basil Bessarion (Diller, 1937: 441-2).

Sophianos’ immensely influential map Totius Graeciae Descriptio, which George Tolias describes as “a visual digest of Greece”, synthesizes two regional maps from Ptolemy’s “Geographia” Tabulae Europae IX and X), contemporary information from portolan (navigation) maps, and narrative and toponymic information from Pausanius’s 2nd-century “Description of Greece”, as well as earlier geographical information from Herodotus and Thucydides (Tolias, 2006: 163) (figure 3).

The question of geographical provenance

Issues surrounding the geographical rather than simply the chronological provenance of Byzantine maps and geographical theories are equally vexing and require us, again, to think beyond the mere location of a map’s original publication. Moreover, the location where the object map is catalogued and archived may be the furthest removed from where it was created: Edson and Savage Smith study the so-called “Astrologer’s Map”, one of the earliest ancient Greek maps of the world (c. end of the 5th-beginning of the 6th c. CE) that was discovered among Greek astrological texts bound in the midst of an Arabic manuscript in the Bodleian Library, Oxford (MS Marsh 42) (Edson and Savage-Smith 2000:7) (figure 4).

In fact, the map’s life-path may have been rather complex. It may have originated in Alexandria, was consulted and – given its symbolic rather than practical nature – used in rituals. And it was possibly owned by a number of patrons across the Mediterranean, until it was collected in an Arab library, and ultimately acquired by an Orientalist scholar who donated it to the Bodleian. Context is frequently critical for determining provenance, although context does not help in the case of the Astrologer’s Map.

Although I would rather not confuse the reader further, I should note that when it comes to the production, distribution, and consumption of Byzantine maps, traditional territorial definitions of the Eastern Roman and Byzantine Empires get in the way. Conventionally, following the Tetrarchy – the Roman Empire under Emperor Diocletian was divided into a Western and an Eastern one. Following the establishment of Constantinian rule, the geographic footprint of the Eastern Empire waxed and waned significantly, assuming its maximal territorial extent under Eastern Roman qua Byzantine rule during the reign of Justinian I (c. 482-565 CE) (figure 5).

Starting with Emperor Heraclius’ reign, the now Hellenized Eastern Empire (575-641 CE) faced increasing territorial challenges from Visigoths and Lombards in the West, Avars and Bulgars in the North, and Saracens and Persians in the East.

Further, the establishment of the Arab Empire in the 7th century initiated a period of great intercourse between Hellenic and Arabic scholarship, inclusive of geography and cartography. After the Arab conquest of Egypt, Syria and Palestine, Ptolemy’s seminal 2nd century CE geographic and cartographic opus becomes the object of systematic study and elaboration by Arab cartographers, astronomers, and mathematicians.

Periodically pushing back successfully against a growing number of geopolitical competitors, as during the reigns of Basil II and John II Komnenos, but more often losing territories to them, by the middle of the 15th c. the territorial extent of the Byzantine state was greatly diminished, comprising essentially the imperial capital city of Constantinople, and a number of fortified exclaves in mainland Greece. Thus, the dynamic movement of political boundaries and frontiers during the eleven centuries of Byzantine political and cultural life, but more importantly, the mobility of scholars and cartographers across political territories in flux, make for unreliable markers of geographic provenance of maps that are generally attributable to Byzantine cartography and geographic science.

Ultimately, the geographies of Byzantine map production, use, and collection point to five contexts: the imperial and Arab administrations in the East, great monastic libraries like those on Mt. Athos and the Monastery of St. Catherine in the Sinai, scholarly collections, private collections of maps of essentially antiquarian value, and merchants and transport concerns that used specialist maps (such as navigational ones). Among all these contexts, it is libraries that at times have been best able to fix in time and space these rare and enigmatic maps.

Fortunately, there are times when geographic provenance could not be clearer: when the map is integrated into the structure of a building, it may endure for hundreds of years beyond the people and the state that produced it. There are at least two such Byzantine maps that are worthy of mention. The mosaic floor map in the church of St. Doumetios (a three-aisled basilica dated to c. 550-575 CE) in Epirus, Greece, is an example of cosmological mapping that interweaves elements of ancient Greek and Christian symbolic geographies. The St. Doumetios mosaic map strains our cartographic imagination (figure 6). As creatures of modernity we would not call that beautiful mosaic a map, since we commonly think of maps as two-dimensional scaled representations of the surface of the Earth. It is also very unlikely that a European map collector of the 16th or 17th century would be able to make sense of this mosaic as map.

For these Early Moderns, cartography was already substantially a science of statecraft, war, mercantile capitalism, and a visualization tool of the new physiographic sciences. But for a 6th century Byzantine person, the mosaic would have been easily identifiable as a map – the Christian cosmos as interpreted through Greek cartographic science. Drawing on the ancient tradition of Anaximander and Eratosthenes (figure 1), the artist-cartographer surrounds the “oecoumene” (the inhabited Earth represented as the world of living being-trees and animals), by a stylized rendition of Oceanus (the impassable ocean that circumscribes the inhabited world, as well as lands in the South, either inhabited by non-human creatures, or devoid of life).

The inscription leaves little to doubt: Kitzinger translates it as “The boundless ocean having in its midst the earth” (Kitzinger, 1973: 370). The 6th c. mosaicist thus marries artistic and aesthetic media of the pre-Christian world, with Christian imagery. Kitzinger continues that “the idea of the earth in the midst of the ocean is conveyed by means of the classical device of an emblema (or panel picture) surrounded by an elaborate frame” (ibid). He likens it to a later, highly decorative example of floor map from the Romanesque Church of St. Salvatore, Turin: “a decorator’s paraphrase of a world map” (Ibid).

Another example of monumental cartography can be found in the contemporaneous mosaic representation of Palestine and the Eastern Mediterranean, located in the apse of the church of Saint George at Madaba, Jordan. That floor mosaic comes closer to our idea of a map, though geographic accuracy is not the artist-cartographer’s objective. Depicting a substantial region that spanned an area from Lebanon to the Nile Delta in the south, and from the Mediterranean Sea in the west to the Syrian desert, the mosaic maps some 150 biblical cities and towns, labeled in Greek. Sites like Jericho and Bethlehem would have been familiar to Christian pilgrims (figure 7). The monumental map of an important part of the Christian world would have been trodden by worshippers, perhaps as a reminder of the mundane character of the physical earth.

Above the mosaic, the Church of St. George (Kitzinger claims it is dedicated to St. John) would have soared over the heads of the worshippers, representing the higher moral sphere (Donner, 1992: 3-5). I would suggest that the Madaba mosaic map can be linked to a particular Christian exegesis of the universe – one that would have been consistent with the Eastern Syrian and Antiochene view of Christianity that influenced another very important Byzantine map, the Christian Topography of Cosmas, the so-called Indicopleustes (Kominko 2005: 164).

In the case of a vastly different object, manuscript MS 1851 from the Codices Vaticani graeci (Biblioteca Vaticana), dated with confidence to the year 1179, we know with certainty the point of origin and destination. Hilsdale describes it as “a small and intimate book whose simplified vernacular voice, large, clear script, and detailed miniature cycle suggest that it was intended for Agnes’s young eyes alone” (Hilsdale 2005: 477). This extraordinary object that contains stylized, landscape-oriented maps of Constantinople, was intended for the Princess Agnes, daughter of Adèle de Champagne and Louis VII of France who wed Alexios, the purple-born son of Emperor Manuel II Komnenos.

The title of Hilsdale’s work (“Constructing a Byzantine Augusta: A Greek Book for a French Bride”) captures and interprets the complexity of design and discourse that define this manuscript. Intended as a manual on Byzantine imperial etiquette as well as an introduction to the Byzantine “topos” over which she was one day to reign, the book is rich in narrative descriptions, visual storyboards, and stylized illustrations of the capital city (figure 8).

What this little book for the future Augusta did was to create a cognitive map of Byzantine imperial life and space. Although the visual vignettes may not be immediately interpreted as “maps” by a modern person, these landscapes have the same effect of fixing Byzantine geographies (the city, landmark buildings, the spatial organization of the palace, as well as the micro-geographies of body posture and ritual precedence in court ceremonies) in word and image for a foreign princess.

In this case, “topography” becomes “cartography,” and finally becomes a princely education. Her actual arrival on imperial lands was at the city of Thessaloniki, where she was celebrated with an oration by Eustathios, former Master of the Rhetors and Archbishop of that city. The oration has been preserved as Madrid MS Esc. Gr. 265 [Y.II.10] fols 368-372 (Real Biblioteca de El Escorial Vol. 2 [Madrid, 1965] pp. 120-131).

As Hilsdale reports, by all indications, her transformation from Agnes, Princess of France into Anna, Byzantine Augusta was successful. Robert of Clari recounts her encounter with French Crusaders in 1203, to who she refused to identify herself as anything but a Byzantine Empress: Clari reported [ele][sic] “ne voloit parler a aus, ainsi I faisoit parler un latimier, et disoit li latimiers qu’ele ne savoit nient de Franchois” (Lauer, 1924: 54, in Hilsdale, 483).

In sum, then, there is often a compelling mystery that surrounds the study of geographic origin and provenance of Byzantine maps. That mystery has to do with the mobility of these objects from the earliest years of the empire but especially following the Latin conquest of Byzantium (1204-1261). Political and military expediency, the vanity and often greed of collectors abroad, the taste for illuminated Byzantine manuscripts, and the desire of Western scholars to access Greek and Arabic classical knowledge produced a massive outflow vectored West. Stories of commerce and conquest may best explain why there are only a few hundred Greek manuscripts in Istanbul today, while the Bibliothèque Nationale, the Biblioteca Apostolica Vaticana, the Biblioteca Laurenziana and the Biblioteca Nazionale dei San Marco, among other Western libraries with Byzantine holdings, catalogue more than 5,200 manuscripts (Nelson, 1995: 209).

At times, emigre Greeks, like Nikolaos Sophianos, would travel to Ottoman-occupied Greece to purchase Byzantine-era manuscripts. Based on a letter by G.B. Amalteo to Paolo Manuzio (27 February 1561) we know that Sophianos traveled to monasteries in Thessaly and Mount Athos between February and September 1543, acquiring some 300 manuscripts for a Spanish collector (Diller, 1937 in Tolias 2006: 172). Tolias further informs us that Sophianos’ manuscript hoard included “mathematical and minor geographical treatises, such as Agathemerus’ Diagnosis and Hypotyposis, and the first lines of Dionysius Byzantius’ navigation of the Bosporus” (Tolias 2006: 152). The complexity of the geographies of Byzantine map production, consumption, and collection speak of a Mediterranean world, at least, that was a highly integrated commercial, security and intellectual space, at least until the consolidation of Ottoman power in the east.

Suggesting a Byzantine map taxonomy

We have already noted the challenges of establishing chronological and geographical provenance in Byzantine map connoisseurship. We will bracket these challenges for now, and focus on the basic taxonomy of maps we accept as either putatively or specifically Byzantine: One important taxonomic variable is ‘scale’ – mappae mundi or world maps constituting an important map class. There are cases where maps at a larger geographical scale (depicting regional detail), such as the later Sophianos map of the Greek world, would be based on a section of a mappa mundi.

Although there is high quality scholarship on specific medieval qua Byzantine map objects, especially since the benighted, and orientalist writings of C. Raymond Beazley at the end of the 19th century, there has been little work done on building a sensible and comprehensive taxonomy of this cartographical heritage, that would allow researchers, including social scientists who are not, for example, trained Byzantinists or papyrologists, to use these objects for a deeper analytical understanding of the Byzantine spaces and their relation to the world.

Equally critical variables are the types of use intended for the map: We can visualize a number of binary or spectral classifications that would describe the map’s original versus derivative character, its symbolic versus empirical nature, its organizational structure, its intended and/or actual use or instrumentality, its situational context, and the media in which the map was produced.

Within the context of mappae mundi, and maps which are derivative from them, we can discern a significant map population distribution across a spectrum: maps of a symbolic, ritualistic qua religious, or magical/astronomical character and uses occupy one side of the spectrum, and maps of a non-religious, pseudo-scientific or scientific, and instrumental character (for example, maps in the service of navigation, travel, location of cities and towns, political-administrative divisions) collectively describe the opposite end.

For the sake of our scientific conversation, I list below some possible pillars of a practical taxonomy:

Original-Derivative: Woodward calls ‘transitional’ maps that incorporate elements of older maps into newer editions or into entirely new maps. Milanesi and others prefer the term ‘synthetic’.

Visual/Graphical-Narrative: Early Christian pilgrim travel accounts are built around lists of directions, distances, and landmarks, not unlike the directions one gets with a Mapquest query today. Per Leyerly, who studied the travel narratives of the Bordeaux Pilgrim, of Egeria, and of the Piacenza Pilgrim, these lists of geographical features and landscape queues were used very much like maps, hence my decision to include them in this taxonomy.

Planned Orientation-Landscape Orientation: Planned views of the earth are conventionally considered ‘maps’, though there is significant support among both historians of cartography and geographers that landscape views (or orientation) constitute an important subclass of maps. One would readily agree with this premise if one has ever used the planispheric orientation tool in Google Earth.

Manuscript-Monumental: Byzantine-era manuscripts of all types (in papyrus, vellum, or other useable medium), number in the few thousands, which would make them rather rare. Among them there is a small number of maps and manuscripts that can be generously classified as maps (as in the case of the landscape maps of Constantinople in the Augusta book). Set apart from these portable maps, are maps integrated into buildings as murals or mosaic floors that can be best described as monumental-the Nikopolis and Madraba mosaic maps being good examples of that subtype.

Symbolic/Religious/Magical/Abstract- Empirical/Instrumental/”Realist”: Maps that resemble the Astrologer’s Map, the subclass of symbolic/cosmological maps we call “T-O” maps (the so-called Pseudo-Isidoran Vatican map (figure 9) being a mature and elaborate version of the type), the St. Doumetius mosaic floor map at Nikopolis, and the Cosmas Indicopleustes map, which draws the world as a representation of the table of Moses’ tabernacle, can be collectively described as abstract cartographic illustrations of biblical exegeses of the World.

Their taxonomic ‘antipodes’ are maps conceived and produced by cartographers, and consumed by clients, who had an interest in the mapping the surface of the planet as accurately as possible (scholar-scientists like Claudius Ptolemy, navigators who lived and died by the accuracy of their portolan maps, merchants whose wealth depended on accurate geographic and logistical information, and strategists, politicians, and administrators (including the propertied institutional Christian Churches of the East and West), who managed territories and vast properties, defended boundaries, and distributed privileges and power that were rooted largely on territorial control (figure 10).

Some preliminary conclusions

We do not know very much about Byzantine maps and Byzantine cartographic science in general. The material record is scant, many objects having been lost, destroyed, removed from their proper context (most often, to the West), or discarded when no longer useful, as is probably the case with Byzantine portolans between the second half of the 13th century and the Fall of Constantinople.

Nevertheless, there is a distinct possibility that more Byzantine manuscripts and specifically maps will be uncovered in libraries with significant historical collections, such as the Vatican Library, the Bodleian at Oxford, the National Library of Greece, and the Ataturk Library.

Yet in spite of the paucity of the material record, there is much that we can discern to build a picture of Byzantine cartography. Its connection to the mathematical, astronomical, and geographic vanguard of the Hellenistic period – the opus of Claudius Ptolemy – is unquestionable. Its basis on geographical models of the pre-Socratic philosophers, and the Classical period historians is quite evident at least in zonal and in T-O type maps. Furthermore, its interest in engaging in a biblical exegesis of the world through the medium of cartography can also be strongly supported. The use and study of landscape as spatial visualization speaks to a certain amplitude in how maps were thought of, constructed and used.

Taken as a whole, we can compose a useful, albeit incomplete picture of Byzantine geographic and cartographic science and practice. As some of the related objects traveled to Venice, Genoa, Spain, the Arab world, and beyond, to inform and influence geographic and cartographic sciences and production there, other objects became part of the scientific and artifactual world of the Ottomans and their subject people. It is reasonable to assume that Byzantine cartography, and the cartographies of Venice and Genoa through commerce and the Latin conquest of Byzantium, influenced the manner elites and everyday people visualized cartographically and modeled spatially their power position within the region we sometimes call the Balkans, within the Ottoman Empire, and in the successor states that replaced it.


Bagrow, Leo. “The Origin of Ptolemy’s Geographia.” Geografiska Annaler, Vol. 27 (1945), pp. 318-387.
Beazley, C. Raymond. “New Light on Some Mediaeval Maps”. The Geographical Journal, Vol. 15, No. 4 (Apr., 1900), pp. 378-389.
________________. “Early Christian Geography.” Transactions of the Royal Historical Society, New Series, Vol. 10 (1896), pp. 85-109.
Chekin, Leonid S. “Easter Tables and the Pseudo-Isidorean Vatican Map.” Imago Mundi, Vol. 51 (1999), pp. 13-23.
Diller, Aubrey. “A Geographical Treatise by Georgius Gemistus Pletho.” Isis, Vol. 27, No. 3 (Nov., 1937), pp. 441-451.
Edson, Evelyn; Savage-Smith, Emilie. “An Astrologer’s Map: A Relic of Late Antiquity.” Imago Mundi, Vol. 52 (2000), pp. 7-29.
Hilsdale, Cecily J. “Constructing a Byzantine Augusta: A Greek Book for a French Bride.” The Art Bulletin, Vol. 87, No. 3 (Sep., 2005), pp. 458-483.
Kitzinger, Ernst. “World Map and Fortune’s Wheel: A Medieval Mosaic Floor in Turin.” Proceedings of the American Philosophical Society, Vol. 117, No. 5 (Oct. 25, 1973), pp. 344-373.
Lauer, Philippe, ed. Robert de Clari, la conqueste de Constantinople. Paris: E. Champion, 1924, p. 54.
Leyerle, Blake. “Landscape as Cartography in Early Christian Pilgrimage Narratives.” Journal of the American Academy of Religion, Vol. 64, No. 1 (Spring 1996), pp. 119- 143.
Maja Kominko. “The Map of Cosmas, the Albi Map, and the Tradition of Ancient Geography”. Mediterranean Historical Review, Vol. 20, No. 2, (2005), pp. 163-186
Milanesi, Marica. “A Forgotten Ptolemy: Harley Codex 3686 in the British Library.” Imago Mundi, Vol. 48 (1996), pp. 43-64.
Nelson, Robert S. “The Italian Appreciation and Appropriation of Illuminated Byzantine Manuscripts, ca. 1200- 1450.” Dumbarton Oaks Papers, Vol. 49, Symposium on Byzantium and the Italians, 13th-15th Centuries (1995), pp. 209-235.
Tolias, George. “Nikolaos Sophianos’s ‘Totius Graeciae Descriptio’: the resources, diffusion, and function of a sixteenth-century antiquarian map of Greece,” Imago Mundi, Vol. 58, no. 2 (2006), pp. 150-182.

Appendix: Illustrations

(Note: all graphics supplied courtesy of the author)

Figure 1a


Figure 1a: Anaximander

Figure 1b


Figure 1b: Eratosthenes

Figure 2


Figure 2: Ptolemy, 2nd cent. ''Map of the World.'' 1482. John Carter Brown Library, Brown University.

Figure 3


Figure 3: Nikolaos Sophianos,Totius Graeciae Descriptio (Rome, Antonio Blado, 1552). Copperplate. 4 sheets. 74x110 cm, in Tolias, 2006: 161.

Figure 4


Figure 4: The Astrologer

Figure 4 (translated)


Translated version of the Astrologer's Map.

Figure 5


Figure 5: Core Byzantine Empire (green during Justinian I

Figure 6


Figure 6:Â St. Doumetius Church mosaic map of the Christian cosmos. There is a reasonable conceptual link between this apparently abstract biblical exegesis of the world, and so-called T-O mappae mundi of the same period. The

Figure 7


Figure 7: Detail of Madaba, Jordan mosaic map, Church of St. George Source: M. Disdero, photographer, Madaba 21/02/2007 (open source). The toponymic richness of the Madaba (some 150 names of cities, towns and other features) would reasonably represent local knowledge. Locals, educated pilgrims and other travelers may have also had knowledge of Ptolemy

Figure 8

2004-2009 Back Archives