The World Map between Workshop and Laboratory

It’s not easy to detect how workshops transmitted protocols of mapping or the content of maps before authoritative practices for recording geographic knowledge  Partly because of the difficulty to describe the transmission of knowledge in manuscript maps, and partly because of the foreign nature of terrestrial continuity in these maps, historians have tied the techniques of engraving with an ability to delineate and effectively demonstrate a geographic space that viewers could process and comprehend.  Tools of engraving afforded a flexible graphic syntax to represent the discoveries of the New World and diffuse the Ptolemaic models of mapping expanse on a uniformly bounded graticule of parallels and meridians from the early sixteenth century, but had earlier afforded a systematic symbolic structure to organize the terrestrial world.  Manuscript world maps based on Ptolemaic cartographical models vaunted their ability to orient viewers on the indices that embodied a continuous bounded expanse.  Yet the inventive abilities associated with the uniform graticule of parallels and meridians gained such demonstrative force in ways that were informed by a rich Mediterranean tradition of nautical charts, often discounted because of the association of engraved maps and modernity.

Yet the nautical charts were adapted in experimental ways as a foundation for assembling a record of terrestrial expanse.  Indeed, charts first registered shifting knowledge of the inhabited world, and a means to symbolize the expanse of an inhabited world:  indeed, the genre of nautical charting first described the inhabited world–the “ecumene“–as a format able to represent and indeed process the settlement of a terrestrial expanse greater than most viewers could imagine.


hmartellus world map


This small “world map” illuminated in Florence about 1490 was been described by Peter Barber as one of a critical map in world history because of how it registered Portuguese travels around the Cape of Good Hope in 1488, and expanded the distance from Lisbon to the coast of China to 230°–expanding its share of the inhabited globe.

Yet rather than use spatial conventions for transcribing expanse, this small map,  adapted the conventions of nautical charts to do so–conventions that note coastal ports, promontories, estuaries, or coastal cities to map the settled world, rather than chart routes of travel.  But this small map, revising Ptolemaic cartographical dogma to accommodate Portuguese descriptions of Africa’s coast after Bartholomew Diaz’s rounding of what became known as the Cape of Good Hope, seems less designed for orientation than it envisions a comprehensive  coverage of the earth’s surface in the reduced format of a sheet about 200 by 120 cm.:  its elegant coloration suggests the reduction of a tradition of globe-making; indeed, Martellus  includes the newly mapped Cape as violating its lower ornamental frame, as if boasting about its greater comprehensiveness as a record of the known world’s surface; ports on the southern tip of Africa, way stations for sailors to southeast Asia, expand the world beyond a classical frame of knowledge.

Although the map is taken as evidence of its suggestion of a maritime route to the East that may have reached Columbus, who would have been encouraged by its apparent expansion of Eurasia to 230°, beyond earlier world-maps and beyond its actual size.  Although the map includes no graticule, a decidedly later map by the same cartographer with a coordinate grid also expanded the coast of Africa beyond the parameters of its frame, as if seeking to accommodate nautical information in an existing format:




Neither map tells us much directly about practices of mapping, but both raise questions about how nautical charts were adapted in deluxe manuscript world maps and about practices of transferring coastal locations in charts to terrestrial maps:  the first smaller image is perhaps more decorative or emblematic, and reveals minimal attention to islands or ocean expanse beyond Eurasia, but it similarly works to shape provisional knowledge of the terrestrial globe from heterogeneous sources.

The development of a uniform graticule of parallels and meridians would be privileged as a graphical form to chart the uniformity of terrestrial expanse.  But its adoption followed sustained experimentation in a tradition of manuscript mapping that is often neglected–but which provided a sort of laboratory for the expansion of a map of the world’s inhabitation, as well as a cartographical craftsmanship in which the mapmaker accommodated information without a set conventions and in makeshift ways.  This post returns attention to how the tradition of charting, associated with workshops, improvised forms to illustrate terrestrial expanse, often with minimal use of graticules of Euclidean derivation.

Manuscripts that synthesized nautical charts maps to fashion descriptions of the settlement of an inhabited world, unlike the oceanic expanse and locations of ports that were noted in nautical charts:  the material construction of such world maps offer stunning evidence of the plastic nature of mapped knowledge and the modern use of charts, by adopting existing cartographical formats to invest new knowledge-claims in maps, both by compiling a corpus of individual maps and redrawing extant cartographical forms, even though they do not follow such a clear protocol of bounding a uniform distribution of terrestrial expanse in the manner of global projections on a rigorous Ptolemaic model of conformal mapping.

The world map that opened the so-called “Medici atlas“–a sequence of eight maps that suggest aims of encompassing terrestrial expanse–reveals how maps acted as compilations of cartographical forms to experiment with recording terrestrial expanse in the same city where Martellus worked and enjoyed wide popularity.


Medicean Atlas


It’s difficult to know how these sequence of nautical maps in a single codex in Florence’s adopt a Genoese nautical chart to present a new model of map making, revising formats of charting widespread in port cities like Genoa to record sea-routes in order to provide a comprehensive pictorial rendering of inhabited space–the world map included not only the cities Ptolemy noted, but cities mentioned by Marco Polo, the Venetian traveller, as the Delhi Sultinate, and texts of the geographer Ibn Battuta, and charts.  It amplified its content through the addition of a sequence of additional maps, also of nautical derivation, of the Italian peninsula before manuscript maps of the Aegean, Black, Adriatic and Caspian Seas, each including cities and the last showing the landlocked closure of the Caspian sea, that oriented viewers to extra-Mediterranean geography with far greater precision than earlier world maps.

As well as revealing interest in the compilation or collation of evidence in maps, its revision of the template of an Africa unable to be circumnavigated and a closed Indian ocean probably responded to considerable interest in late fifteenth-century Genoa in regions of “Africa nondum cognita,” suggesting a clear interest in new working practices of mapmaking.  Alexander von Humboldt, who knew the map well, took it as evidence of a medieval knowledge of Africa’s circumnavigation–so difficult was it for him to separate the map from a register of what was known.

Perhaps the expansion of Africa to the Cape of Good Hope was drawn atop the existing map when the true form of Africa was known, or even added to the original chart in ways that emulated later prototypes, and in essence kept the cartographical compilation “up to date”:  if the case, suggested by the coloring of the map beyond an inked outline of the continent, it would suggest a new use of the map as a canvas that was altered at the same time as transmitted in fixed protocols or techniques.  It would also suggest a process of re-visioning the inhabited world by revising the pictorial content of the map to foreground its descriptive abilities.

Why was the manuscript “atlas” made, what readers did it address, and how did it construe geographical expanse?  Did it try to consciously convert the inhabitation of coasts so emphasized in portolan charts to descriptive and representational ends?

Is it representative of the expansion of the tradition of charting to newly communicative ends as a demonstration of worldly settlement?  The term “atlas” was devised by Mercator to describe a collection of maps, and its application to this volume of maps is, strictly speaking, anachronistic–the collection of maps are widely referred to by the term used by the Dutch cartographer and engraver Abraham Ortelius, under Mercator’s influence, in 1570 to describe his collection of maps, the Theatrum Orbis Terrarum, as an atlas–the compilation of maps exploited the comparative use of nautical charts to organize a record of cosmographical scope of comprehending the world’s expanse.  But the term is appropriate to describe this codex of maps, since it captures the persuasive ends of presenting a sequence of nautical maps of terrestrial expanse to offer claims of complete geographic coverage.

The world maps showing Africa as surrounded by sea predates by almost a century Portuguese circumnavigation of its cape.  The distinctive world map in the “Laurentian atlas”  clearly suggests the navigability around the continent and passage from the Atlantic ocean to Indian ocean–and considerable knowledge about the shape of Africa for an atlas of eight sheets dated 1351, but probably composed around 1370.  It is difficult to judge whether the map was drawn in its current fashion at that date–a time when few sailing maps extended south of Sierra Leone–or whether the original map of a Genoese chart-maker was redrawn, as some have argued, to comprehend later Portuguese nautical discoveries in the Atlantic and Indian Ocean.  The African coastline in the map included no toponymy on the coastline below Cape Bojador, the southernmost point of most earlier voyages.  Was an earlier chart revised, better to comprehend shifts in terrestrial space, earlier than the Martellus map pictured above, to better incorporate new nautical knowledge by extending the reach of Africa beyond Cape Bojador?


Painted Expansion of Atlas?


The map designed in the format recalling a chart and without any attempt to include indices of measurement or orientation may reflect the considerable interest in routes of trading to the East independent from land-travel. But also staked broader claims as a “world map” of terrestrial coverage in persuasive ways, evident in how it expanded the known limits of the continent to create a new image of inhabited expanse–using the chart as a canvas in ways that reflected some early fifteenth-century interest in assembling a map of the inhabited world.


Medicean Atlas


In place of the format of a nautical chart, the “world map” suggests a bounded expanse and form of distinctly modern appearance that has often disconcerted several readers.  (Charles de la Roncière wondered with bemusement at the prescience “de la forme reelle de l’Afrique avant le periple Portuguais” in the world map.)   It seems to have been redrawn in the fifteenth century to incorporate Portuguese discoveries, to register recent discoveries of alternate rendering of Africa’s distinctly modern form, beyond an ink line that the paint conceals.  Indeed,  no toponyms appear on the map south of Cape Bojador in the Sahara–the southernmost point of most nautical charts–in ways that seem to confirm the imagined form of Africa within the chart.  And the prominent location of  some of the Canary and the Azores in the Atlantic reveal subsequent alteration of the map on the basis of Portuguese charts, as if to update and revise its original form.  The revision of the genre of nautical mapping for new ends of comprehensive terrestrial coverage–in this case, by repainting its very surface–suggests a redesign of familiar formats of charting for new claims of geographic coverage to make a point about terrestrial expanse as the first translations of Ptolemaic treatises advanced a model of cosmographic–as much as cartographical–expertise.

The world map is clearly a compilation of varied forms of knowledge and geographic thought.  Despite the vaunted limitations on geographic knowledge of the later middle ages, this mid-fourteenth century world map reveals a striking synthesis of geographical knowledge to illustrate terrestrial expanse.  The volume seems to purposely compile existing nautical charts in a single volume, in the manner of an atlas, to satisfy demand for a record of global comprehensiveness; this map probably drew on Arab sources for its southward extension of eastern Africa down the Muslim Swahili coast.   Some emendation of the map is evident, moreover, detectable in the competing hands of an upright humanist script and more straggling cursive east of the Caspian Sea.  A nautical chart of Italy immediately follows the world chart in the codex, a map focussing only on its coast to the exclusion of its interior save the most prominent lakes:


ITaly as Nautical Compilation


This portolan chart suggests, however, a distinctly terrestrial subject, as much as a compilation of sea-routes.  And as a synthesis of the sequence of six nautical charts that complete the volume, its collation or compilation of charts is uniquely modern in presenting a comprehensive geographic record.  Its comprehensive view of the continent echoes the early fifteenth-century world map of Albertin de Virga, dated 1411-15, whose totalistic content it precedes in its synthesis of nautical charts to depict the inhabited world’s expanse–and which could have offered a model for its emendation–shortly after the first translation of Ptolemy’s manual of world-mapping in Rome.  Did Ptolemy’s newly translated treatise prompt such an early synthesis of available mapping forms to describe terrestrial–rather than nautical–expanse?




The newly persuasive nature of assembling nautical charts presented something of a canvas to stake global relations.  This world map presented sites for the exchange of meaning, but also a visual laboratory to configure terrestrial space,  transferring coastal lines from portolan charts to prepare open spaces of terrestrial expanse to be surveyed.

We often cast nautical charts as both impressive and regressive in contrast to terrestrial maps, because they leave the interior topographies so bare.  The inclusion of a more detailed African coast from the beautiful Portuguese Dourado portolan charts of the later sixteenth-century, for example, drafted in Goa around 1570, whose thickly-packed toponyms hug the continent’s north-west coast, listing ports and riverine mouths which seem the only areas able to be mapped or of interest.  It leaves the terrestrial interior unknown in ways that recall Joseph Conrad’s alter-ego Marlow, and d his attraction to the unknown African interior  in “The Heart of Darkness.”


Dourado Portolan, NorthWest Africa


The chart reveals something of an incommensurability between a known shoreline of coastal settlement and an interior, that was oddly perpetuated through the 19th century, where its limited notion of terrestrial coverage gained colonialist ends.  For printed maps of the mapping of continent limited to coastal African ports s to emptiness of the mapped interior–without any acknowledgement of local toponomy–in maps created the sort of “large, empty spaces” Marlow claimed to have seen, in maps “from the latest authorities” drafted in London in 1805, or the Lincoln & Edmonds 1819 map of the continent, both of which might have inspired Marlow’s voyage to Colonel Kurtz:

Africa Map 1805 %22From The Latest Authorities%22

Lincoln & Edmonds 1819 Africa Map


But if the chart is seen as a conservative force in world mapping, it was also a fertile tradition of its own.  Indeed, the notation of parallels and meridians in portolan charts popular from the middle of the sixteenth century in Europe suggests a clear concern in mapping new lands as well as encounters with coasts–much because they fit less clearly with mapmakers’ expertise.

This is evident, to switch subjects of cartographical attention, in the mapped image transmitted in Portuguese portolan charts of the coast of Brazil, similarly striking because of how it foregrounds the remove of coastal cities or hydrographic records of the Amazon from knowledge of inhabitation of the continent’s interior.  This nautical chart accommodated the practice of mapping on a graticule of parallels and meridians, even if comparable imaginative license is taken with the topography of the interior.


Duarado Brazil close-up


Indeed, the second of the seventeen maps in this ‘Dourado’ portolan map an interior strikingly similar to the lake on which Sir Walter Ralegh mapped the imagined city of El Dorado on the imaginary Lake Manoa in 1595, as fed by multiple freshwater ingresses; Ralegh’s map reveals less fantasy in projecting the voyage to El Dorado than a recycling of the very sailing charts he consulted in drawing up his plans to sail to Guiana.  A comparison suggests the staying power of the tradition of the Spanish and Portuguese nautical charts Ralegh so frequently consulted to plot his ill-fated expedition to an imagined city he detailed for viewers at the very center of his chart:



Ralegh’s re-use of nautical charts  created something of a laboratory to assemble the means to envision terrestrial expanse in the early fifteenth century Mediterranean world.

The portolan was, in multiple ways, a particularly versatile and inventive form.  It maintained a similar inventiveness across cultures in the Mediterranean, adopted into a tradition of Ottoman mapping through Portuguese prototypes.  Much as the ‘Medici atlas’ suggests a new mapping of the oceanic enclosure of Africa, the re-use of nautical charts for cosmographical ends is evident in the below image of world expanse.  The map was used to illustrate a Renaissance manuscript of the thirteenth-century cosmographer, lawyer and physicians Al-Qazwini’s “Wonders of Creatures and Strangeness of Beings” [Aǧāʾib al-maḫlūqāt wa-ġarāʾib al-mawǧūdāt], also stored in the Biblioteca Laurenziana.  The map below retains conventions like a cinnabar Red Sea sketched below the Holy Land, spider-like lakes and ports crowded inland that seem to perpendicularly hug the shore, the description of tripartite continents of Asia, Europe, and Africa below testify to its terrestrial coverage.  Scattered trees, fill unknown lands and almost random rows of mountainous barriers mark the limits of known Africa–an cosmographical record of  the inhabited three continents of the world Al-Qazwini had described in the tradition of classical geography, rooted in describing terrestrial inhabitations, rather than climactic zones or schematic spatial divisions–notwithstanding the depiction of a bright crimson Red Sea.




Geography meets cosmography in this world map, deriving from nautical charts.  The image of planetary eclipses in the same elegantly illustrated volume suggests a similar cosmographical intent in this deluxe compendium, that echoes the sense of the terrestrial map as a mediation of abstractly purified knowledge, now removed from the craft of the nautical chart-maker’s craft.




The cosmographical scope of the image in the thirteenth-century encyclopedia re-used the format and conventions of nautical charting to make cosmographical claims, similarly to the re-drawing of the Genoese chart in the mid-fifteenth century for making newly comprehensive claims.

Indeed, one can detect something of a similar use of records of nautical charting for cosmographical mapping in the adoption of Portuguese charts in the fragment of a 1513 map that the Atlantic the Ottoman admiral Piri Reis presented to the Ottoman Sultan Selim I in 1517.  The fragment discovered in the shelves of an Istanbul’s  Topkapi palace in 1929, drawn on gazelle parchment, incorporated four Portuguese portolan charts, eight Ptolemaic maps, and some twenty mappae mundi, according to the admiral’s inscription, in a luxurious map of expansive proportions, of which only the sheet depicting the New World and western Africa survives:

Piri Reis

In the decades after its discovery, this map was taken as grounds for grandiose claims about pre-Columban geographic knowledge–claims with limited validity.  For Piri, the map’s surface provided a field to combine an image of global expanse in a complete whole.  He worked, he tells us “from eight Jaferyas of that kind [e.g., Ptolemaic maps, which he dated to the age of Alexander the Great] and one Arabic map of Hind [India], and “from four newly drawn Portuguese maps which show the countries of Sind [Pakistan], Hind and Çin [China] [all] geometrically drawn, and also from a map drawn by Qulūnbū [Columbus] in the western region” or islands of the new world, so that by “reducing all these maps to one scale this final form was arrived at, so that this map of these lands is regarded by seamen as accurate and as reliable as the accuracy and reliability of the Seven Seas.”  He took the testimony of seamen as the standard of trust, but the value of a format “geometrically drawn” suggested both its comprehensive and cosmographical value–a main purpose, no doubt, of the presentation map given as a luxury book to Sultan Selim I–making it all the more quixotic that the director of the palace library, Halil Edhem, handed it over amidst a bundle of discarded materials to  Gustav Adolf Deissmann when Deissamn searched for non-Islamic material in the Topkapi.  (So suggestive is the map’s situation of islands and accumulation of detailed coastal promontories that it was taken as grounds for a lost “pre-classical” culture of mariners familiar with Antarctica for Charles Hapgood or of the mathematical calculation an azimuthal projections made with active extra-terrestrial assistance by Eric von Däniken from a spaceship located above Cairo; theories that misread the map as an actual projection had little basis, but are nicely rebuked by the cogent case made for its nautical origins on the Turkish 10 million lira banknote.


(The fragment of a lost manuscript map depicted to the left gained such wide traction precisely because of its elegant  argument for ordering the known regions of the world.)

The nautical chart provided by its very open-ness a basis to expand a canvas of the inhabited world before conventions of map projection insisted on the world’s bounded nature.  The flexibility of the format allowed the assimilation of a varied range of sources and authorities–from Marco Polo to Ibn Battuta, as it were.  We can appreciate the open-ness that the re-use of nautical charts as canvases in contrast to the very different nature of the world map in another manuscript of Al Qazwini’s treatise, now in the National Library of Medicine, painted in around  1537  by unknown illuminators in western India.  The map is far more schematic–if not traditional–form, removed from the tradition of charting:  the image of the world is replete with marine sinuses and undulating rivers that link bubble-like lakes, with a bulbous Horn of Africa pointing to the overlapping waves in the oceanic expanse of the Arabian sea.

al qwazwini  1537 w india


The vitality of a manuscript tradition of mapping provided something of a laboratory for the material re-imagining of an earthly space in which increasing physical details were readily inscribed.

Leave a comment

Filed under Africa is a Continent, Mappaemundi, Mapping Africa, mapping the ecumene, Nautical Charts, portolan charts, protocols of mapping

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s