Tag Archives: World Maps

Monuments in New Worlds: Placing Columbus in the Americas

Commemorating Christopher Columbus’ transatlantic voyages in a “discourse of discovery” has been magnified as they have become a bit of an exercise of collective remembering and of world-making. For Columbus monuments are less markers of fifteenth century voyages, than they serve to frame a range of narratives of discovery that promote the fifteenth-century navigator as an icon of nationhood that were foreign to the fifteenth century. In making claims for the foundational role that the navigator’s transatlantic voyage, they create a new narrative of nation, particularly powerful for its ability to occlude and obscure other narratives, and indeed the presence of local inhabitants in a region, so that they assume the deracinating violence of a map: as claims of possession, and indeed mastery over space, they dislodge nativist presence in a region, much as Columbus did as a royal agent, glorifyinf the acts of renaming, and taking possession of, the new world to ally the viewers with the heroism of the Genoese navigator’s foundational act of taking possession of the New World.

Columbus set the basis for such grandiose claims, perhaps, by staking the transatlantic “colony” La Navidad as the first settlement by Europeans of the New World on Christmas. The event’s commemoration has continued in statuary across the Americas, even if the thirty-nine settlers left behind in the settlement were all massacred by natives after their ship, the Santa Maria, was dismantled after it hit a sandbar, and the men of the lost caravel that ran ashore on a sandbar off the island Columbus had auspiciously renamed Hispaniola, as evidence of Hispanic claims to sovereignty for his patrons, its hull and crossbeams converted to a fortress that was later burned to the ground. While the crew left with a translator who were instructed to collect a store of New World gold as they awaited his return may have been overzealous in courting natives that they were all killed, the extraction of riches and wealth and slaves motivated the New World settlement.

Yet in the commemoration of Columbus Day, the landfall has been renarrated as an inspired revelation of new lands, and allied with a sense of manifest destiny in retrospect, foregrounding the imperial mission of laying claims to new lands under the eyes of God. Yet the elevation of Columbus as a figure of state, and a westward course of empire, was magnified only in the nineteenth century. Paradigmatically, the painting of Columbus taking stock of his arrival by Dióscoro Teófilo Puebla Tolín, staged before uncomprehending natives cowering offstage, occurs as a missionary raises a cross on verdant shores where sailors triumphantly raised recognizable standards of the Spanish sovereign in the New World.

Dióscoro Teófilo Puebla Tolín (1832-1902),
First landing of Columbus on the shores of the New World, at San Salvador, W.I., Oct. 12th 1492,

re-rendered by Currier and Ives, 1892/Library of Congress

The bucolic image of arrival was cast as a triumph of technology, civilization, and deliverance. The bucolic scene not only denied violence, but was an image of paradisal promise that Dióscoro Teófilo Puebla Tolín rendered in a manner widely reprinted in the lithograph of the commemorations of quadricentennial celebrations of 1892 as a narrative of westward expansion of Manifest Destiny. The recent re-questioning of Columbian commemoration as a common national identity led to the questioning of commemorative Columbian statuary across the United States, from Columbus, Ohio to San Francisco to Kenosha, WI, to Miami. As the statues were dislodged from the common memories of an Italian-American community–as many once were in New Haven, Boston, and Philadelphia as well as New York City–their place loosened in a narrative of nation in ways that needs to be told.

Attracted by a remarkable burst of creative iconoclastic energy, San Francisco’s City Arts Commission preemptively monument to Columbus somewhat preposterously overlooking the Pacific to be removed from its monumental pedestal in order to maintain the local peace–a statue long defaced in recent years–before it was defaced. The removal of monuments to Columbus spread as a re-tallying of moral accounts, but a restoration of civic peace. The importance of refiguring the commemoration of colonization grew as a form of reparations whose logic was unmistakably national in character, if the first questioning of Columbus Day had been local and selective in 1992-3. The deposition of the 4,000 pound statue, with a violence that would repeat and channel the rejection of the figure of Columbus whose monuments were already deposed in Boston, St. Paul, Minnesota; Camden, NJ; Richmond, VA, and other cities in New York state, one of which was beheaded–if long after the statue to the navigator was ceremoniously pushed into the ocean in 1986, in Port-au-Prince, Haiti, with a placard “Foreigners out of Haiti!” And by 2020, the ease with which statues of Christopher Columbus were assimilated to the confederate monuments preserved across many southern states–and preserved at considerable cost to American taxpayers–reminded us of how easily the memory of Columbus as colonizer was cultivated among white supremacists as iconic testaments to a sense of historical security of another era we were trying to pry ourselves free in hopes to gain distance on.

Even if the statues of Columbus and Columbus commemorations are overshadowed by the razing of the statues of Confederate generals in recent years–

Southern Vision Alliance,
Confederate Monuments Removed since George Floyd’s Murder

–the images of the dicsoverer were dismantled as we engaged a contested legacy. If monuments removed with an eye to toppling racism across southern states that had commemorated secession in the attempt to defend enslavement and chattel slavery were a stain on the nation that emerged like a return of the repressed in the summer after George Floyd’s murder by overly zealous “law enforcement” forces, the removal of monuments had begun as undeniable evidence of their talismanic status as lodestones for white supremacy became clear after violence in Charlottesville directed attention to the degree to which commemoration of the Confederacy kept a memory alive in national and local consciousness, revealing how undecisive the Civil War was for the preservation of local memories across many border states or secessionist states, and the toxic nature of preserving memories of southern secession as a defense of what were cast as local liberties within the union.

The division assessment of historical legacies that shape a narrative that informs the present landscape of inequity had been contested for decades around the heroicization of the figure of Columbus as a shared national point of reference. As much as the seven hundred and eleven standing monuments of commemorating secession–over 1500 statues which are collectively preserved by taxpayers’ money at a cost of $40 million for annual upkeep. The standing statues dedicated to anti-abolitionist figures have kept the memory of the Confederacy alive across the United States, including of Jefferson Davis and Robert E. Lee in Virginia, Georgia, North Carolina, and Texas, creating a topography that has inflected political identities in ways Donald Trump was savvy if hateful to tap, as if to present or re-present an incubus already planting seeds across the land, many of which were only removed by energized (and disgusted) protestors or whose remove was ordered by city councils in an attempt to preserve the local peace–including of the unidentified heroic “Confederate soldier” who removes his hat in downtown Alexandria, VA, removed only in June 2020–if similar statues remain in Jacksonville FL, and San Antonio TX, heroizing as if sanctioning the very option of local resistance to according rights to enslaved populations, beyond preserving memories of war dead.

Mapping the hundreds of Confederate statues across the US | Black Lives  Matter News | Al Jazeera
Southern Poverty Law Center/Al Jazeera graphic

The confederate statues were long mythologized as an alternative system of justice, echoing the reduction of rights and civil equality across the landscape by holding up a distorting mirror of Southern victory and secessionist pride, gaining legs as grounds to advocate an outdated status quo reflective and constitutive of an alternative order unto its own, as is evident in the naming of courts of law after Confederate names.

Visualizing a Confederate Present, 2017/Loretta C. Duckworth/Scholars Studio/Temple University

The spattering of blood on Columbus memorials call for a revision of public memory complicit in a culture of spatial colonization that perpetuates the fundamental nature of racial hierarchies. The requestioning of commemorating the act of violence as fundamental to the nation’s values is a questioning of their place in civil society, and their meaning to the nation, motivated in no small part by the ugliness with which they have been seized with new currency as images justifying a racial superiority and sanctioning the violence of enslavement along racial lines. For their removal had tried to call attention to the dangers of commemoration by targeting the figure of Christopher Columbus, whose statues had first multiplied across American land in roughly the same era of the later nineteenth century, following Confederate statues, in a sort of monument trick that served to naturalize white possession of indigenous lands.

The overturning of commemorative statues of the fifteenth-century navigator so deeply dissonant to our sense of national belonging, common memories, redressed disturbingly long-lasting spatialities–the average statue was almost a hundred years old–as the nation entered a temporal loop of recursive nature of reparative bent, as the destiny Columbus imagined for himself as a civilizer and discoverer of a New World–and new continent–emerged in increasingly pressing ways, opening up the very speech act of taking possession of the Americas as a fiction, only masquerading for utilitarian ends as a binding legal precedent. For only by confronting the painfully exclusionary nature of such an act of taking possession, deriving more from the practices of enslavement and mastery of others that run against the very basis of our own civil society, or the civil society we seek to create.

Owen Thomas, San Francisco Chronicle

Indeed, the San Francisco’s 4,000 pound commemorative statue of Columbus, often defaced as a symbol of enslavement and subjugation in recent years, was removed by a crane and as a call to dump it into the Bay was circulating, on Thursday, June 18, removing it from a scenic site by the Pacific beside Coit Tower, leaving an empty pedestal, perhaps to reduce the need to clean up a statue that had been repeatedly defaced in recent weeks but also to show consensus about lack of interest in defending a symbol of oppression, enslavement, and colonial violence, and public outbreaks around the call to depose the statue off Pier 31, not as a symbol of colonial resistance, but an expunging of the navigator from national history. All of a sudden the dismantling of public memories of Columbus’ heroism were national news, a divisive issue responded to not with understanding but professed shock for besmirching American history, not reassessment of values, battling Italian-Americans Nancy Pelosi herself as forsaking, as if to bemoan her betrayal of the preservation of the hallowed memory of Christopher Columbus in the summer of 2020 to stoke lines of political division in the heat of the 2020 Presidential campaign: the fate of the statuary of Columbus was a bell whistle for stoking fears of a danger to the status quo.

It was as if the spontaneous prominence across the nation of memorials to George Floyd, proliferating on street walls in full color, and in haunting offset likenesses, provoked introspection demanded introspection of what sort of memorials we identified with and wanted to see the nation, placing on the front burner of all the question of commemoration in terms that had long been glossed over and tacitly accepted. The questioning of commemoration after Floyd’s murder came to articulate a spontaneous rebuke of the continued validation of racialized policing and police violence, throwing into relief discriminatory monuments. There were soon few defenders of the monument able to tolerate how they emblematized division of the social order, eager to ask us to situate Columbus more broadly rather than historicize his complicity in “some of his acts, which nobody would support,” without addressing the framing of the logic of “discovery'” in imperial narratives of conquest and disenfranchisement of indigenous claims to sovereignty and to recognize the need for reparations.

For the navigator embodied an imperial relation to space and terrestrial expanse, discounting the inhabitants of regions, and affirming the abstract authority of sovereign claims and sovereign expanse, however improbably early maps placed the islands in the Caribbean–later called Hispaniola–based on his conviction that the Atlantic Ocean was able to be traversed, enabling transatlantic voyages for which Spain was well poised to expand commerce far beyond the coast of Africa and the Mediterranean for economic ends in an “Enterprise of the Indies” that Columbus proposed to John II of Portugal, before he set out to claim the new lands for Ferdinand and Isabella. The longstanding embedded nature of Columbus in a discourse of claiming land–a discourse from which he was not only inseparable, but embedded maps in claims of the administration and supervision of lands far removed from seats of terrestrial power, a map-trick that has been celebrated since as a form of inscribing territorial claims on a piece of paper or globe.

And if Columbus had no actual idea of the form of North America, the persuasiveness of fictive reimagining of his mastery over space–a mastery cast almost uniformly in intellectual terms, rather than in military terms of disenfranchisement or enslavement–provided a logic that is aestheticized in the monument as a mode for the possession and persuasion of possession over terrestrial space more akin to American hemispheric sovereignty in its open heroizing of a national geopolitics of the 1890s than to a Renaissance discourse of discovery–comparable to the reimagining of hemispheric sovereignty in the years after the 1867 withdrawal of Spanish sovereignty to Mexico.

The origins of these reframing are perhaps obscure, but lionizing Columbus was always about rewriting the American narrative, and distancing one race of immigrants–the Italian migrant–from the very native inhabitants that the story of Columbus displaced. The navigator was promoted actively as a figure of national unity in the post-Civil War centenary of 1892, in which Columbus assumed new currency as a national figure, a map on silver able to enter broad circulation as a memory for how a three-masted caravel mastered terrestrial expanse, resting above a hemispheric map of global oceanic expanse. The anachronistic map suggests as much a modern triumph of hemispheric cartography–the coastline of the United States was surveyed by geodetic terms and that established the role of the United States Coast and Geodetic Survey in producing maps of uniform toponymy and hydrographic accuracy had only recently set standards of coastal surveying that unified triangulation, physical geodesy, leveling, and magnetic of authority within the US Navy to produce coastal maps of the nation extended from the Gulf of Mexico to the Alaskan shoreline.

The imperious gaze of the limp-haired navigator seems the first self-made man as he gazes with gruff determination on the coin’s face, almost entirely filing the surface of the first American coin bearing human likeness. Columbus was an icon it identified with how the hemispheric map took charge over a continent, and gave a sense of predestination to the recently settled question of continental integrity–and a territorial bounds that new no frontier up to Alaska, whose coast had been recently surveyed, and much of Florida and the Gulf of Mexico. Its design for the Chicago Word Exposition suggest a hemispheric dominance reflecting the growth of Rand McNally in Chicago, a map-publisher for America, as well as the self-assertion the United States as a hemispheric power, as much as the Genoese navigator about whom so many meanings have encrusted.

The striking hemispheric map of global navigability on the obverse of the coin circulated in Chicago’s World Exposition was global, but would also mimic the claims of hemispheric dominance that the hemispheric projection recalled, prefigured the Pan Am logo, in its global in reach–as if the image of a spherical projection devised by Rand McNally that spanned the globe and erased all borders might be cast as the seedbed for globalization was itself contained in the transatlantic voyages of the small trade ships, the Pinta, Niña and Santa Maria that were led with hopes of a profitable economic voyage with Columbus at the helm. (Rand McNally had not only sponsored the world’s fair, but its double spherical projection that recalled Columbus’ conviction of a spherical world by ahistorically featuring a cartographic design Columbus would have known; the planar projection was an icon of global expansion and conquest, more detailed in coverage than late seventeenth century double spherical projections.

–but devised and issued its own elegant version of a world map based on the Mercator projection in following years from 1895, in atlases issued subsequent to the World’s Fair, to meet a growing market for global maps. Leaving much of the African interior unmapped, in a manner that cannot but recall Conrad’s Heart of Darkness, the image is a confirmation and announcement of global triumph, centered on the North American continent and United States, if it shows the world.

Rand McNally Global Map based on the Mercator Projection (Chicago and New York 1895)
Continue reading

Leave a comment

Filed under American history, Columbus, commemoration, Voyage of Discovery, whiteness

Mapping Land and Sea in Venice (and Elsewhere), ca. 1500

The medium of single-line engraving provided an expressive medium for organizing the continuity synthetic maps of land and sea long before trans-Atlantic travel was available to most.  Mapping beyond one’s place or region is a specific area of expertise; it is not surprising that it is a difficult competency to define.  It’s long been observed that the manner in which engraving produced an exactly replicable visual statement brought a variety of levels of expertise to bear on the map, both as a repository of collective visual memory and a coherent visual statement designed to orient readers to the notion of a uniform space.  But it’s interesting to consider the local differences in how a coextensive notion of space was understood to be composed of mapping the integration of land and sea:  and the understanding of the political power of the Serenissima–and the authority of the Venetian senate–as extending “onto the salted waters [sopra le acque salse]”–suggested a unique model of imagining worldly rule that uniquely inflected the construction of a cartographical space.

The transmission of the concept of a map of uniform coverage–one first expressed by the second-century Greek astrologer and mathematician Claudios Ptolemaios [Κλαύδιος Πτολεμαῖος], bequeathed to us as simply Ptolemy, provided a template to illustrate the expanded edges of the inhabited world in the editions successively translated in early modern Europe considerably before maps of land were fully integrated with maps of sea.  Translation of the forms by which Ptolemy mapped an inhabited terrestrial expanse not only to superseded the inhabited world as Ptolemy had described it and imagined it, but broached a different model of continuity within visual form:  epistemologically distinct spaces of travel that corresponded to different forms of mapping were joined in a Ptolemaic planisphere, as were the distinct competencies of mapping, in what might be profitably examined and studied as a ‘trading zone’ of varied forms of technical skill.  In Venice, perhaps, more than in other sites, the city offered a site connected by sea to other regions, and was a bit of an active trading zone of linking maps of water and land as Ptolemaic maps attempted, less constrained by territorial bounds, and more attentive to unifying the different metrics and scales of global mapping than many other engravers of global maps.

The image of the Venetian ties to the Mediterranean world and Gulf of Venice were rearticulated in maps, ordering a relation to a global expanse by nautical charts that had enjoyed broad currency in the city, and provided a cosmographic authority to articulate Ptolemy’s authority as a description of a terraqueous expanse.

 

1.  Mapping Land and Sea

Techniques of artistic engraving offered a matrix in which to synthesize mapping forms from the fifteenth century, and the medium increased a synthesis of formats of mapping, as well as a the demand for maps as reproducable forms.  As much as benefiting from Ivins’ useful characterization of the innovative ways that print afforded “exactly reproducable graphic statement,” engravers’ skills provided a way to transmit the map as a graphic form.

The Dutch engraver and cartographer Mercator in 1569 described his map as a synthesis of geographical maps and nautical charts:  in so doing, he modernized the projection of the map’s surface as a continuous surface.  In a unique and inventive way, Mercator assembled a record of terraqueous expanse on parallels and meridians to address a large audience of readers by boasting of his ability to bridge the distinct media of nautical charts of the ocean with geographic maps–whereas the Ortelian “Typus Orbis Terrarum” of 1570 directly below displayed traveling ships, riverine networks, and maritime expanse on curved meridians, Mercator’s projection distributes an inhabited expanse on perpendicularly intersecting meridians and parallels.  But both maps advance cartographical expertise as preparing a surface that could be uniformly scanned by viewers as a proportional and uniform distribution of the inhabited world.

OrteliusWorldMap1570

Mercator Close-Up

 

Mercator did not explain the mathematics of a uniformly mapped space.   But the unique projection he devised gained broad authority by the seventeenth century as a means to visualize global relations.  Although the Mercator projection ensured that the loxodromic lines of nautical travel, denoted in charts by rhumb lines, would perfectly intersect with meridians, the straight parallels construed “ad Usum Navigantium Emandate Accomodate” was not adopted for sailing or for plotting voyages until the ability to measure longitude at sea–partly since he did not explain his method for calculating the “true course” on straight lines, but also since the media of terrestrial maps were so distinct from nautical or navigational carts.  But the combination of registers for noting nautical and terrestrial space, or imagining expanse on ship and on land, provided a major shift in maps’ graphic design and epistemological claims.

The gradual supersession of the autonomy of  sea-charts facilitated increased claims of realistic representation–or reality effect of mapping land and sea in a continuous frame of reference.  The combination of geographic and nautical charts to record of the known world in ways shifted how the world is known depended on acceptance of the descriptive potential of maps, as much as their accuracy or the use-value they gained to navigate in an era when calculation of latitude at sea depended on the sighting the altitude of the sun above the horizon and due course rarely achieved.  But the Mercator projection integrates land charts and marine charts to provide totality of global expanse.  This was the first age of globalism, and it could be readily understood.  The cognitive basis of maps as vehicles seems concealed in Cornelius de Jode’s presentation of Mercator’s projection as a “Totius orbis cogniti Universalis descriptio” or record of the known world in 1589, a decade after its appearance:  it offered tools for knowing the inhabited world as well as a record of the known world.

A similar visibility of the world’s surface was advanced in Cornelius de Jode’s later compendium of global coverage, which synthesized the conventions of nautical charts with the conventions of terrestrial mapping to create a convincing understanding of relationships between nautical travel and terrestrial expanse.

MErcator 1579

Such supersession of the conventions of mapping had to an extent previously occurred in the combination of results from different mapping formats in a unified cartographical space.  Yet even before Mercator devised this projection, sixteenth-century maps had synthesized the content nautical maps had increased the claims of realistic representation–or reality effect–in printed maps.  The graphic and pictorial detail and abundance of signifiers that was invested in Ptolemaic projections had increasingly shifted the status of the map from a schematic register which lay at remove of one from space, to a compelling synthesis of terrestrial relations for its viewer:  and the map became a surrogate able to stand in metonymic relation to a place it described, stood at the center of  modern claims of maps as forms of visual relation to space that could be inscribed with meaning.  Indeed, the combination of registers of terrestrial and nautical cartography compellingly joined areas of practice that had been kept previously separate formats of spatial descriptions, if not incommensurable registers of qualitatively different registers to chart spatial continuity.

Was this change in attitudes to the map partly enabled by the combination of registers of terrestrial and nautical cartography?   From previously separate formats of spatial descriptions, if not more significantly incommensurable registers of qualitatively different forms of space, the map’s surface became understood as a way to register motion through a uniform space and encounter places on a determined path of travel.  The use and status of the map as both a register and descriptor of expanse was already evident in the integration of nautical charts and Ptolemaic maps Bernardus Sylvanus designed in Venice after 1508, which, despite the notorious absence of coverage of North America and reduced coverage of the globe, had previously described a terraqueous unity in compelling–and readable–ways by explicitly combining what had been seen as incommensurable orders of registering expanse.

Sylvanus PLanisphere--close up

The legibility of the 1511 projection of Sylvanus was both very contingent and local in nature, despite the universalizing viewpoint it prepared of the inhabited world.  When Sylvanus, hailing from Eboli, possibly also an illuminator of erudite texts, undertook a newly illustrated printed edition of the ancient geographer Claudius Ptolemy’s  Guide to World-Mapping, known in the Renaissance simply as the Geography, he decided to create a more updated edition of comparative maps deriving from nautical charts collated by sailors and the set of maps transmitted in codices of the Ptolemy’s work of global geography.  The plans for a new Venetian edition had recently been abandoned, although several plates for it had been made and perhaps engraved by 1508, probably including a new world projection.  In confronting problems of modernizing the Ptolemaic maps, Sylvanus foregrounded the integration of islands and coastlines compiled in nautical charts in the maps transmitted in Ptolemy’s geographical treatise, translating the conventions for land-mapping into representational conventions from the graphic arts and advances of two-color typography:  the birds perched on the cornices of his map of Italy, the sixth plate of Europe, may echo the modern bird’s-eye view of the peninsula he offered, using nautical maps to present the configuration with a sense of naturalism often foreign to early printed Ptolemaic maps.

Sylvanus Italy--Europe 6

2.  Oceanic Space

The treatise that the second-century geographer titled a “handbook for drawing world maps” was both a technical guide and a compendium for drafting land maps.  But in Venice, a city of maritime trade, Ptolemy’s promise to collate and list a database of all the places in the inhabited world’s surface had potential appeal as incommensurate with the chart used to decide or compare nautical routes of travel, and posed a specific challenge to synthesize mapping forms.

These charts provided an alternate source of information that promised both to refine and expand the ancient geographer’s encyclopedic claims that led him to list names of ancient cities and noteworthy cities or rivers exceeding 10,000 in number–if the richness of Ptolemy’s text led erudite readers to consult his book with their manuscripts of Herodotus or Livy, as Bernardo Machiavelli–father of Niccolò–their elegant terrestrial maps they more often addressed learned readers and armchair travelers as surfaces often read in relation to other ancient texts, rather than graphic descriptions of expanse.  Indeed, their printers did not aim to address a larger audience of readers.  Yet even when presenting accurate place-locations in coastlines that resemble charts, the maps struggled to offer an easily readable surface.

sugar hillls in Spain

The synthesis of a more legible cartographical space was foreign to earlier cartographical traditions.  The history of the transmission of medieval maps is considerably complex–as are the techniques of varied forms of map making.  Elizabeth Edson argued information from accounts of travelers, traders, and sailors became accomodated in world-maps from the early fifteenth century, joining travelogues that both expanded the content and challenged the parameters of earlier symbolic world maps.  The inclusion of information from travel accounts and nautical charts not only expanded the surface of maps, but posed complex problems of integration on parallels and meridians–a reproducible grid–and elicited potential graphic models for spatial representation over a century after its textual translation that lent formal authority to the world map.

The alternatives for such a synthesis were not clear.  The considerable questions that surround the transmission and construction of earlier manuscript charts, often drawn on sheepskin to guarantee their preservation and illustrate their value, are raised by the unclear relations between how the maps were transmitted and copied–if not created–given the unclear questions about copyists reliance on the intersecting directional lines that seemed sketched over their content in tracing coastal shorelines and locating islands, or how the skein of lines apparently determined from compass-bearings provided guides for nautical travel.  These maps were produced predominantly in port towns, as this Mediterranean chart executed in Alexandria by Jehuda Abenzara (or ben Zara), coastlines are crowded by names of coastal ports written perpendicularly to the shore, linked by a network or web of potential sea-routes that demand close reading and intense preparation by specially trained scribes:

Jehada Abenzara

In port cities like Alexandria, chart-makers regularly synthesized and collated a sort of collective memory of varied routes of travel that might be on board any arriving ship, in the hope of piecing together these local records of coasts or island-charts to synthesize more expansive networks of trade with a degree of accuracy that minimized cartographical distortion with a precision that geodetic observations had not allowed.

The chart synthesized a form of collective memory, if the protocols by which its contents were transmitted are not clear:  the organization of a synthetic record of travels provided little more than symbolic reference to inhabited interiors, however, which in essence remained “off the map.”  Rather than a representation of terrestrial space, it primarily provided a record of the location of ports and idealized potential lines of nautical–rather than terrestrial– travel.

mostra-cartografia

The spatial mapping of coastal cities in the Mediterranean, and situation of coastlines in a broad nautical expanse–both in relation to both equinoctial lines and vertical bars of latitude however provided an alternate orientation to the network of the web of loxodromic lines of the compass rose.  The below schematic version of a portolan chart, signed by Juan de la Cosa of c. 1500, provided a distinct frame of reference and spatial indices to enumerate points of landing and prominent capes in the New World at different latitudes for its readers.

Wrote de la Cosa's c 1500 map

1500_map_by_Juan_de_la_Cosa

 

The parchment portolan chart stored in Madrid’s Museo Naval and made in the port city of Andalusia, Puerto de Santa María, was prepared for competencies of a restricted audience, with specific interpretive tools in mind–whether they were kept by captains, or by trading houses is unclear, as is the primary techniques they use to demonstrate relations of space.  By the fifteenth century, elegantly decorated versions became prized possessions among even landlocked elites–probably in copies that obscured or hid their own mercantile provenance and were designed to stake boundary lines of exploration or colonization in the New World, by demonstrating the boundary line of Tordesillas.  But although the competencies of mapping these documents enlist to render expanse are opaque, their synthetic construction have provoked continued investigation of their formal manipulation or symbolic construction of mapped space.

Some of the relevant underlying schema of the networks and constellations in charts have been identified, but their operative value is not known–were they of use for copyists in Salamanca, Barcelona, or Genoa, or were these keys that allowed them to be read?  The construction of scale lay in the relation among focal circles, wind roses, and loxodromic lines, as in this reading of the Cantino Chart.

800px-Compass_grid_Cantino_planisphere_(1502)

Spatial position is not much of an apparent interest, however, so much as the collation of alternative networks of travel–or, in the case of some charts presented by the Spanish or Portuguese, to illustrate the meridian that demarcated colonization of the New World at the Treaty of Tordesillas.  The image of nautical continuity was a huge attraction for the humanist geographer Martin Waldseemüller, but his 1516 “Carta Marina” based on Portuguese marine charts like the so-called Cantino chart constituted part of his broader cosmographical project, but this image, discovered only by the Jesuit Josef Fischer around 1901, constituted an alternate model of cosmographical learning to his large world map of 1507, 4.5 to 8 feet, provided a wall-map whose comprehensive character was less successful in making claims for its legibility, if it invested greater artistic skills in converting the format of nautical charting to a legible form that Waldeseemüller had the projection engraved in the same dimensions.  This map printed on high-quality hand-made rag paper was only found in one sixteenth-century bound volume, but was a complicated investment, even more so than the cosmographical map that Waldseemüller described as having been printed in 1,000 copies.

Carta-Marina-LG

 

Somewhat oddly, the map did not include the image of “America” surrounded by oceanic waters that distinguished the lavish cosmographic wall-map he had printed in 1507, and whose accompanying treatise described America as “an island . . . surrounded on all sides by sea,”  in his Cosmographiae Introductiomost probably because its sheets reflected the content of sea-charts–even if it superimposed an equi-angular grid that had little relation to the graticule employed in the terrestrial wall-map he had titled a Universalis Cosmographia.

 

Oceanus Occidenatils

 

The two large wall-maps produced at the University of Vosges, then in the Holy Roman Empire, both only recently acquired and restored by the Library of Congress, enshrined opposed if  incommensurable models of world-geography at the very time Sylvanus prepared his own edition of Ptolemy’s precepts of geographic map-making and study of global geography.   Did the lavishly produced “Carta Marina” offer a counterpart to the geographic theorization of expanse that Waldseemüller had advanced in his cosmographical writings?  The ordering of Venice’s position in relation to a gulf, and to the expansive genre of island books or isolari printed in Venice and in Italy, provided a new way of describing Venice’s position in the world, and global continuity at a relatively early age.

 

3.  Envisioning the Continuity of Terrestrial Geography

The location of geographical in the continuous coastlines of manuscript nautical charts was hastened by a demand to process the over 12000 identified sites Ptolemy specified as able to be mapped in a format  which conformed to viewers’ expectations for representing spatial continuity.  And Sylvanus seems to confront this difference shift in collating nautical charts with other mapping forms in Venice around 1510,  in what seems a uniquely local manner to read a map’s universal claims.

The detailed coverage of the world’s surface in sixteenth-century Europe increased not only the coverage or precision of maps, I would argue, so much as the claims of realistic representation–or reality effect–of maps in critical ways.  Yet changing understanding of the map as a medium, as well, provided Bernardus Sylvanus with grist to collate nautical charts in a set of new conventions that created a uniformity among data of diverse provenance previously regarded as qualitatively distinct if not incommensurable orders of spatial description.  Although his exacting transposition of ancient names into modern outlines of land-masses ran against the critical project of comparing the ancient and modern worlds, the uniform conventions of maps he made presented a distinctly uniform continuous surface in images from charts.

For charts were less concerned with describing or denoting spatial location, than determining (and collating) potential routes of travel:  the conceptual mapping of routes of travel was rarely invested with descriptive force or value; its competency reflected applied knowledge.  The growing authority of the terrestrial map as a comprehensive description, however–one of the deepest of modern claims of maps as competencies rooted in visual design, rather than nautical knowledge–arose from the combination of registers of terrestrial and nautical cartography, previously separate formats of spatial descriptions if not more significantly incommensurable registers, in a sort of a trading zones of semiotic conventions from varied areas of life, which bridged or linked hitherto incommensurable formats to denote expanse.

As the rich spatial information contained within the medium of the chart was transposed to the surface of terrestrial maps,  something like a wrestling with epistemological claims for knowing space and locations seems apparent in the maps included in treatises of global geography first translated in the fifteenth-century, most particularly in Claudius Ptolemy’s second-century Guide for Drawing Terrestrial Maps, whose maps Renaissance editors of the treatise had increasingly invested with increasingly comprehensive ends–increasingly relying on the toponymically crowded but crisply defined coastlines transmitted in charts to blend seamlessly with inland areas.  The accumulation of local and pictorial detail to combine an over-abundance of signifiers altered the distinction between the land map and nautical chart, raising truth-value claims about the chart as a representation that stood at remove of one from the world that this post can only begin to suggest:  increasingly, the map became a place that could be inscribed with meaning, or became a register from which to relate to foreign lands, if not a substitute for them.  The diminishing authority of the chart lay partly in a limited ability to determine position at sea, but also a limitation of the ability for encoding further information in its content that would satisfy its audience.  Edward Wright observed the errors of sea-charts as a basis for calculating position in 1599:

EdwardWright-CertaineErrorsinNavigation-1599

A word or two about this complex treatise, abundantly overflowing with strange toponyms that elicited readers’ curiosity even if its content were difficult to translate into the standards of eloquent expression to which many of its humanist readers were habituated–leading some to indicate Strabo as–to quote Isaac Casaubon–the “summo scriptore, quod praeter acuratissimam totius orbis nunc cogniti descriptionem, tanta doctrina, tamque varia omnium rerum scientia refertum est, ea denique arte contextum . . .

Ptolemy’s expansive catalogue of locations had long demanded to be given a visual form.  The question of their visual coherence led some of his later editors to rely on nautical charts that included places Ptolemy had not indicated, but the nautical chart provided little analogous framework of coherence by which to grasp their situation in a continuous expanse.  The geographer Angeliki Tsorlini has recently employed digital technologies to map relative locations defined by the terrestrial coordinates in Ptolemy’s treatise in ways that reveal the very compelling map of Mediterranean cities his treatise would have offered.  Most of the cities are ports, located along the shore, to be sure, but a considerable number remain inland cities located with apparent relative precision, with minimal significant distortion for much of Italy, the Adriatic, and Greece.  The copious abundance of familiar locations and interest in their clustering must have increased demand for their depiction.

Place Names from Ptolemy in Modern Map Projection

In the first codex that arrived in Rome, found by Maximous Planudes in the late fourteenth or early fifteenth century, the abstract ordering of the situation and topography did not pose an intellectual problem of viewing space (Burney 111; British Library).  Despite the formal appearance of the island of Taprobana, thought to perhaps represent Sri Lanka, the red lines of parallels of latitudes and meridians of longitude in which Ptolemy argued geographic mapmakers could usefully divide the world for readers on measured units, provided limited claims to mediate a naturalistic image of expanse.

Maximous Planoudes' Taprobana

Planoudes was careful to note the precise location of places on spatial coordinates, but the metric values of locations were not presented as lying in exact correspondence to their spatial situation.  The illustration of cartographical images that expanded later codices of Ptolemy’s treatise worked hard to provide maps that were commensurate with the over 1200 place-names–including mouths of rivers, promontories, mountains, or landmarks–contained in his geographic compendia were sought to be illustrated in authoritative form.

As the work reached a large audience in manuscript, terrestrial space was presented in schematic terms, the maps seem to wrestle with the abstraction of space, as if in ways that could not be imagined in visual or pictorial terms as a surface that could be scanned, as is evident in this map of German lands in one codex of the Geography, which enumerated towns and rivers in a new abstract form, listing inhabitants and towns as in the Ptolemaic manner, with minimal recognizable guides or explicit orientational clues about their spatial situation and topographical location, even when that region lay on the margins of the Roman world:
Magna Germania forests in Swabia

Yet the land-locked nature of these regions made the legibility of expanse less concrete.

Even in areas that claimed continuity with the ancient world, the production of Ptolemaic treatises curiously included modern views of Mediterranean cities in several deluxe of codices illuminated in Florence, as if to expand the treatise’s qualitative coverage of European cities in a rhetorically persuasive image for readers–these images had less regard for the systematic terrestrial coordinates Ptolemy proscribed than for preserving noteworthy sites in each place, or offer a ‘chorographic’ complement to Ptolemy’s explicitly geographic concern.

996179_671379542888235_530184269_n

4.  Symbolic Syntheses of Mapped Space

The question of what sort of graphic synthesis was provided in a geographic map is broadly tied to Renaissance visual culture, but posed particularly pressing questions in port cities that compared Ptolemy’s precepts with maps of nautical expanse.  Bernardus Sylvanus assembled engraved maps for his edition of Ptolemy shortly after the plans to print an edition of the treatise in Venice collapsed or failed, for reasons of skill or financing.  But a huge shift occurred in the production of maps that made such authoritative regional claims as depictions had already occurred, reflected in the preponderance of their incision, illumination, and distribution in centers of visual cultures in northern and central Italy, central Germany, and the Netherlands:  the specific forms of overlap between nautical and terrestrial methods in sites from Venice to Rome to Nuremberg created a rich repertory of maps with expansive truth-claims as forms of depiction.  His work came on the heals of an existing experimentation with combining cartographical registers of description in a universal register of mapping habitations of terrestrial space, evident in the 1507-8 world map of the Roman edition of Ptolemy, designed by the northern engraver Johannes Ruysch, and contemporary to the plans for a Venetian edition of Ptolemy’s treatise.

The manner that this 1507 world map mediated the legibility terraqueous expanse as a continuous surface might have offered a model for Sylvanus’  integrating of mapping forms:  for the Ruysch projection is in ways a restatement of cartographic expertise.

Rome 1507

 

The black-and-white outlines of the copperplate incision helps foreground the legibility of toponyms and textual panels alike that lie on the map’s curved meridian lines, as the stippled surface of oceanic expanse suggests the fact of its comprehension in the map–a comprehension rendered evident to viewers by the unveiling of the new form of a circumnavigable Africa and India, as well as the introduction of the newly discovered capes, rivers, and islands of the Americas:

 

Tolomeo-Stampa-Roma

The historian of cartography David Woodward argued that cartographical competence reveals a growing “rationalization of space” around 1492.  In ways, we have begun to remove cartography from a professional genealogy that places a premium on rationality–such a claim is concealed within the creative combination of forms of diverse sources mapmakers have long imaginatively integrated in synthetic designs.  But the limitations on the ‘rationality’ of the map–or the grounding of its authority in its rationality–demands future research for how mapmakers who amplified the local qualitative content of cartographical media.

Taking a step further backward in time, we can perhaps appreciate how the designers and illuminators of maps of maps included in manuscripts of Ptolemy’s treatise seem ambivalent in their use of parallels and meridians as a framework for defining a cognitive relation to expanse or for recording a cognitive relation to place:  for they treat the graticule of the map more as a frame of reference by which to register terrestrial position, than as an enabling format for graphic representation:  the iconic portrayal of place in early maps as clusters of houses that positioned against the blank ‘space’ framed by coordinate system or patches of forest tries to bridge Ptolemy’s ancient model for denoting a uniform abstraction of terrestrial expanse on Euclidean precepts and the ability to transcribe space.  Illuminators, few of whom are known,  invested maps with very limited mimetic qualities from the 1450s and 1470s to communicate their continuity:   the new interest in regional maps as registers lead illuminators to position clusters of houses with peaked roofs and taller towers in dense proximity to each other to distinguish areas of settlement, beside clustered areas of forest growth–as the Black Forest in Bavaria–that provided some vague reassurance of the correspondence of space.  Some of the owners of such maps added places near their own residence, or areas that they knew, omitted in the printed editions or codices they owned, as if to give the maps an expressive value that they feel they lacked.

Added cities of Hamburg and Lubek

Bohemia in 1477 Ptolemy

 
Did the second-century geographer’s “handbook for drawing world maps” have different implications in Venice, a city of maritime trade and considerable diversity, where nautical maps were more prevalent than maps of terrestrial expanse by the early sixteenth century?
 
 
4.  Back to Bernardus Sylvanus in Venice, ca. 1500
 
The shift in Venetian culture for locating place in a map’s expanse is reflected in the collation of a set of independent views of neighborhoods to create a dramatic imagined synthetic view of Venice as seen from above in a wall-map composed from six large individual woodblocks and large rag sheets.  The master of perspective Jacopo de’ Barbari designed the detailed view by taking he city’s coasts a a frame in which to distribute its built and inhabited expanse:  heads of winds of each direction frame the view, recalling the spokes of a wind-rose and the disembodied heads of putti who surround most early printed Ptolemaic maps, magnify the city’s coastlines and maritime surroundings, revealing the complexity of its physical plant as if the city were something of a microcosm of the inhabited world, and to showcase the expansive position of Venice on the Adriatic.  The view situates the “forma urbis” not only as a built space but in realtion to the surrounding sea, dotted with individual boats and a regatta:  in the distance, one sees the Alps to the north:  the city appears as a microcosm of global expanse, as the depiction of its inhabitation in each rione of Venice stands as a graphic surrogate for the mapping of a miniature world.
 
 
Jacopo_de'_Barbari_-_Plan_of_Venice_-_WGA01270
 
 
The particular detailing of a sea as continuous with coastlines and inhabited world provides the informed viewer with something of a metaphor for the unity of land and sea in world-mapping, revealed in Jacopo’s attention to both wind-heads round the city and to a regatta that braves Adriatic winds, exploiting his attention to the finely engraved lines of the wavy waters:
 
 
Barbari Regata
 
 
What sort of view did Jacopo de’ Barbari compose in this elegant multi-sheet wall map?  The view is often compared to the elevated “bird’s-eye” perspectival views of the “forma urbis” of Renaissance cities, but rests on a synthsesis of an imagiend or virtual view from individual surveys of the city:  one recent digitization of the view of “Venetia 1500” helped reveal the synthetic unity Jacopo took pains to created a uniformity of urban space from individual surveys as an illustration of considerable skill of rendering an almost planimetric space for viewers to scan as a continuous surface that extended to the surrounding oceanic sea:
 
 
Gridded view of Jacopo's Venice
 
 
The multi-sheet map, whose production required three years, exemplifies a Venetian appreciation of elevating a record of collective perceptions by combining map-making and perspective with particular virtuosity.
 
Jacopo_de'_Barbari_-_Plan_of_Venice_-_WGA01270
 
 
Each of the six sheets provided detailed records of the city in what Fortini Brown has called an “eye-witness style,” but a imported mapping records to a continuous picture-frame that pushed the cartographic metaphor of transcription to transcend a single fixed perspective.
 
 
Barbari Close-Up with Tritone
 
 
The luxury print of multiple sheets provide a surface into which the viewer can descend into specific neighborhoods or regions that are immediately recognizable:  the continuity of its content were thematized in another recent digitization of the map created by the Correr Museum:
 
But the lines of the Venetian lagoon and Adriatic suggest the clearest inclusion of a sense of maritime space in the map–an illusion that was echoed in the corpus of Sylvanus maps.  For Jacobo de’ Barbari created a model for viewing the coherence of urban space that responded to a challenge for ordering the unity of terrestrial and nautical space.  When Bernardus Sylvanus intended to expand the cartographical corpus of Ptolemy’s Geography in Venice around 1508, he consciously and proudly incorporated information from the surface of sailors’ nautical charts into the land-maps denoted by spatial coordinates in earlier editions of Ptolemy’s treatise, creating a unified legible cartographical surface and using printer’s red to place cities in a continuous landscape–if often situating ancient names of place from Ptolemy’s work within the modern coastlines of nautical charts, in ways that went against the scholarly tradition of comparing ancient and modern geography by juxtaposing “ancient” and “modern” maps, but also advanced a single cartographic record as authoritative and unique, shading coastlines to suggest the maritime field in which he placed new nautical discoveries–and limited America, famously, to the Columban islands to the ahistorical exclusion of all North America.
 

219-v1-800x600

Rather than enabling spatial travel, the world map of two sheets noted place-names in a distinctive printer’s red that stand out from rolling hills, framed by etched lines of waters on their coasts as if in imagined relief:

Sylvanus expanded Mediterranean with nuatical maps

The map’s space was treated as a continuous surface, defined by the coastlines from modern nautical charts, if the toponomy was often ancient in origin, treating the cartographical surface as a uniform register of inhabited lands:

Sylvanus Spain Coast

 

Little biographic information is known about the production of the maps of Bernardus Sylvanus da Ebola, though he has been possibly identified with an illuminator.  But he clearly exploited, even more than his predecessors, the semiotic synthesis that print allowed in Venice.  This is evident both in its combination of text and woodcut imagery in this two-sheet map, and the overlay of a graticule, equatorial bar, and wind-heads, combining conventions of different mapping media more explicitly than even earlier editions of the existing maps of the Ptolemaic corpus.

The introduction of islands and coastlines not in most all of the maps editors of the previous five printed editions of Ptolemy’s treatise on world-mapping (a sudden burst of editions which we can label Bologna 1477; Rome 1478; Ulm 1482; Berlinghieri 1482; and Rome 1507), presenting more clearly identified coasts and islands–as the ‘isole fortunate’ off of Africa’s western coast, although it omits the New World–but are often of limited geographic accuracy. The distinct use of type to balance the legibility of a map crowded with toponymy by two-color ink adopts the innovation of the material production of books to create a surface easily read by its customers–and he invited readers of the maps he organized as a comparison between the maps Ptolemy described and the versions corrected by modern nautical charts to “compare Ptolemy’s words with navigations themselves” and decide for themselves, using two-color printing to facilitate an intensive reading of the map’s surface, and in the attention that he gave to islands in the Mediterranean, as the Balearic islands off the coast of Spain, where the etching of lines suggest the surrounding seas that hit their rocky shores.

Balearics

 

The significance of the line in the medium of engraving has been argued to facilitate the conventions of uniform mapping of terrestrial expanse, allowing engravers to exploit the geometric formats of Ptolemaic mapping in graphic form in particularly expressive ways, the expressive value of the Sylvanus maps derived from their synthesis of conventions of map-making in a continuously readable form–one that created new attentiveness, indeed, to the encryption of information from the surface of the map, both in the map of the world’s surface and the individual tables editors helped prepare for Ptolemy’s treatise.

This must have responded to an increase in what might be called geographic curiosity.

The universal coverage of the maps Sylvanus prepared for Ptolemy’s manual of global geography was constructed from a very local place, and reveals the local availability of island books or isolari in Venice, as well as nautical records of the Mediterranean and Adriatic that were available in abundance in the maritime city, which were carefully integrated within the system of parallels, meridians and equinoctial lines for readers to pour over, with attention to areas like Spain’s Mediterranean coast or Greek islands in the Adriatic, depicted by a similar accuracy reminiscent of charts, as are its inlets and bays.

Greek Island Sylvanus

Sylvanus illustrated the division between Africa and Asia, the origins of the  Nile and shores of the newly-mapped Red Sea for readers to consult, probably in relation to available maps, by means of a similar etching of graphic relief:

Africa and Origins of Nile

The material surface of Bernardus’ maps synthesized a range of semiotic conventions that viewers would have been quick to recognize as a combination of a material landscape and a map:  one of his Italian readers was quick to include images of the towns in the Marches in the map of Italy and the Adriatic, depicting both the towns of Monterubbiano and Moresco i in ways comparable to the iconic perspective views of cities.

Sylvanus' Adriatic

The additions suggest a dramatic increased in the graphic materiality of the map as a pictorial register.  Print are allowed men as Bernardus or fellow-engravers and editors of maps in Florence, Rome, and Antwerp to invest the map’s surface with new claims of legibility as a reproducible record.  But it is also very possible that Bernardus’ sustained engagement with a project of printing he hoped would be far more successful derived from the prominent status maps already enjoyed in other visual media.

The interest of maps as depictions reflected a deep appropriation of Ptolemy’s instructions to his own second-century contemporaries to craft a map “ad oculorum aspectum commensurabilis“–the transmission of this precept to later mapmakers to create a surface that would appeal to their readers’ eyes, if not also the tacit presuppositions for viewing a continuous space in a detailed and harmonious form.

5.  A tradition of fifteenth-century Venetian cartographers had incorporated nautical charts to illustrative or pictorial ends in inventive ways, in attempts to give greater expressivity and comprehensiveness to the Ptolemaic planisphere or nautical chart:   a 1448 world map designed with great care by Giovanni Leardo framed by the months of the year and astrological signs (Verona, Bibl. Civ., Ms 3119); Fra Mauro’s famous circular map uniquely synthesized Portuguese charts, a unique matter given that it was in fact commissioned for Portugal’s monarch, without a graticule; it recalls an ellipsoid world map of 1457 constructed on the principles “of cosmographers” without a uniform graticule, and filled with textual legends, fanciful iconography, and perspective city views.  None privileged the geometrical order of a uniformly continuous surface or a format of projection from terrestrial cartography, however, or bridged different semantic registers in the manner of Sylvanus’ maps.

The Ptolemaic model provided an authoritative basis to fashion a surface that could be readily scanned as a uniform distribution of expanse by around 1500, and in Venice shifted the attitudes of viewers to mapped space.  By the later fifteenth century, the Venetian Senate had commissioned the repainting of territorial maps of the lagoon of Antonio de’ Leonardi from his nephew Sebastian along parallels and meridians by “Ptolemy’s doctrine” that Isabella d’Este and others Isabella d’Este sent painters to copy, marvelling at its proportions and scale.  The painted map received praise as “così perfetta nelle sue misure [so elegant and well-proportioned]” that “diversi Principi [several princes]” had commissioned copies of it for their own enjoyment and pleasure before its 1577 destruction, Sansovino boasted among his catalogue of the city’s artistic treasures.[i]

Although the map is now destroyed, and cannot be pictured, it constituted something of a model for the multiple maps now present in the Palazzo Ducale, painted to replace it, and for the maps of the Veneto that Christoforo da Sorte created in its private chambers–as well as, perhaps, Egnazio Danti’s monumental remapping of the peninsula in colored paint.  The much-admired peninsular map may have provided a model for integrating the format of nautical charts with maps of geographic content by men like Sebastian Cabot, piloto for the Casa de la Contratacion in Seville who created a new world map–or the map-engraver and engineer Giacomo Gastaldi, who from 1546 synthesized multiple elegant wall-maps that refined cartographical expertise; Gastaldi’s work with the geographer Giovan Battista Ramusio led him to design comprehensively detailed pictorial wall-maps as that of South-East Asia.

gastaldi 1548

Gastaldi-prat of Asia

But we might also start from the 1511 modern map of the peninsula that Sylvanus designed:

Sylvanus Sexta Tabula with ms addition of city views

 

Did this lost expansive painted map of the lagoon that extended to the Adriatic and Tyrrhenian sea, and their islands, provide a model for uniting terrestrial and nautical maps that men such as Bernardus Sylvanus sought to generalize for a larger audience in printed form?  The reader of Sylvanus’ printed maps from Fermo sought to make the text his very own, adding his own qualitative views of the cities that he knew, in ways that register a distinct relation to the map as a continuous surface. 

When the great cartographer Vincenzo Coronelli mapped the geographical situation of Venice in a broader gulf from the mid-seventeenth century, he described the place of Venice in the expanded gulf in his 1688 global atlas–placing Venice in relation across the Gulf of the Adriatic to the islands of its empire, which bordered on the expanding Ottoman by shifting boundaries, as if to affirm its own domain of the seas that opened along its shores.

Golfo di Venezia

As if overseeing an expanse that might be translated into varied scales, the dominion of Venice was defined across maritime expanse, not by territorial bounds, but in the cartouche from which the emblem of the lion of San Marco serenely oversaw its content.  From the margins of the map, the winged lion that Coronelli cleverly located in the cartouche that looked over the expanse of the Gulf, overseeing the expanse from beneath its dogal crown, beneath six bars of scale of mapping that alligned each of five standards with maritime leagues.

 

overseeing

[i] Gallo, “Le mappe geographiche del Palazzo Ducale di Venezia,” Archivio Veneto ser. V, 32 (1943): 47-54. Sansovino, Venetia, citta nobilissima et singolare (Venice:  Iacomo Sansovino, 1581), fol. 122, “era una tavola d’Italia così perfetta nelle sue misure, che diversi Principi ne domandarono l’essemplare.”

1 Comment

Filed under engraved maps, globalism, portolan charts, Renaissance engraving, two-color typography

Java La Grande

The map “Java La Grande,” an imagined continent that invited close inspection from viewers, gives new meaning to the assertion that the map is the territory.  For although this territory never existed, it was mapped: the medieval construction of the antipodes, a mode to balance the continents, was an artifact of early modern European cartography of considerable duration and lasting power.  When Matteo Ricci mapped the world’s regions in 1607 for his Chinese hosts, his copy of the Ortelian map included a large discussion of “Guinea” as a place not studied or known well by Europeans, and perhaps an independent island or attached to the southern pole–albeit one that was not known to be inhabited, and with which little if any contact had been made.  In this sense, Guinea remained an outlier in the age of discoveries, and a somewhat rare blank spot on Ricci’s global map.

 

Ricci SE Asia.png

 

As an artifact with an esteemed cartographical lineage, the image of the continental landmass continued to be mapped into existence by the mid-nineteenth century.  But its appearance in the below chromolithograph of a Renaissance map from the library of baronet Sir Thomas Philips was declared the first map ever recorded of nothing less than the very continent of Australia, as ostensibly witnessed by the Portuguese sailors in the pristine state of nature in which they had allegedly first discovered it.  The map was prized as an image of initial discovery.

 

1280px-Australia_first_map

 

The anonymous artist-cartographer recorded the recession of space in this map, which as it recedes its inhabitants are registered in brownish tracery, foregrounded the idyllic life of this entirely unknown land.  You get the idea of arriving in a region taken as Australia that was the entrance to a landscape of untold wonders in the map of “Java La Grande” designed for Nicolas Vallard’s nautical atlas of 1547 in Dieppe, a center for the diffusion of Portuguese nautical charts for an elite audience of European nobility, reproduced in this 1856 facsimile now stored among the jewels of the Australian National Library.

The pictorial landscape chart was a luxury embellishment of the sorts of rutters that were drawn with artists of local topography.  The material object was prized and promoted by Phillips as a record of the first encounter with the territory of Australia, reproduced with such painstaking care in the mid-nineteenth century  as a chromolithograph from the first baronet’s library when he tried to disperse the Vallard atlas and other early modern books to the British Library, inspiring the reproduction of the map, which he mis-titled “The First Map of Australia,” to promote the value of his collection   The map made for the Vallard atlas by an unknown painter was probably valued because it seems to record an image of Portuguese encounter with the indigenous inhabitants of Australia.  As part of the corpus of Dieppe charts that were, unlike earlier nautical maps, pictorial syntheses Portuguese discoveries of almost ethnographic qualitative richness as depictions of inhabitants for elite audiences, the image has provided a touchstone of a vanished native culture that is particularly powerful as an imagined precursor of the discovery of Australia, valued as an image of imagined contact.  The Vallard map possesses the distinctively attractive qualitative lushness peculiar even to the Dieppe school of which it forms part.

The identification of Australia with the mythical “Java La Grande” is not entirely rooted in geographic fancy.  Java La Grande was described by Marco Polo as the largest island in the world–but reflected some recognition of this unknown landmass that extended to Antarctica, and recurs in the Dieppe maps as a cosmographical idée fixe as not an island, but terra firma:  at the same time as the current-Java was described by geographers of the 1540s, the Dieppe school identified La Grande Jave as an extension of the Antarctic Terra Australis, and it was taken as an early evidence of the southern continent’s early discovery.

The map became taken as the territory.  And so why not preserve this map as a treasure of the collections of the Australian National Library, even if it is a nineteenth-century copy–and quite a faithful one–of a map that represented quite a different imagined land?

 

350px-Australia_first_map

Early identified as the first map made of the continent, this map provided a list of coastal ports that an artist proceeded to fill in with imagined views of inhabitants advancing as if to greet newly arriving visitors with arms literally open in welcoming signs, if not in a ritual procession.  The intensity of play of imagination is evident in contrast to a later 1777 nautical map showing two ships’ circumnavigation of the world under the command of Antoine de Bougainville, whose cartographer constrained himself to parts of the Eastern coast observed with sufficient detail to establish for readers, allowing far more local detail to the as yet unsettled continent, and confined himself to the better-known shorelines of its coast:

CBA-2030-full

 

Yet what was later prized as the “first map of Australia” is  distinct from a roughly contemporary 1543 chart by Guillaume Brouscon of the same school in Dieppe, who synthesized information from prized Portuguese charts to a new audience of landlocked European nobles.  This earlier map devoted a considerable space to “Java La Grande,”  extending the rectangular format of the engraved global projections later standardized after their printing by Mercator and Ortelius, if already adopted by humanist geographers, and offering far more detailed depictions of the settled interior than nautical charts that confine themselves to coastal towns.  The almost ornamental multiplication of compass roses that proliferate like heraldic crests on this map suggest its ornamental nature in a corpus of maps.

A fair amount of extension of the topos of cartographical invention or the mediation of new discoveries that animated these excited atlases of the late 1540s in Dieppe are reflected the map of Java le Grand among the 56 maps that the nautical sailor and self-styled cosmographer Guillaume le Testu included in his Comosgraphie Universelle, selon les Navigateurs, tant anciens que modernes.  The work’s comprehensive claims derived from its use of a range of Spanish and Portuguese charts together with maps of his own design that synthesized recent maps of the Americas:

 

481px-Le_Testu_GRANDE_JAVE

 

He readily presented this map as rife with cartographical invention, as well as following cartographical conventions, as if suggesting the frequent embellishment of persuasive or recognizable detail in maps, more than the license his own achievements in mapping much of the Americas may have merited:  “what I have marked and depicted is only by imagination, and I have not noted or remarked on any of the commodities or incommodities of the place, nor its mountains, rivers or other things; for there has never yet been any man who has made a certain discovery of it.”  The absence of “certain discovery” is an odd juxtaposition with his own discoveries, and the admission of the absence of such “certain discovery ” led to a land that was entitled to be created by the imagination.

The imagined land’s expanse was documented as well in nautical maps limited to shorelines, offering far less qualitative local details than the expanse of its coast, but suggesting in an enticing fashion its expanse, and multiplying elegant compass-roses as if for an excuse to include more gold leaf, as a sophisticated ornamental boundary of decorative motifs:

 

1098px-Guillaume_Brouscon._World_chart,_which_includes_America_and_a_large_Terra_Java_(Australia)._HM_46._PORTOLAN_ATLAS_and_NAUTICAL_ALMANAC._France,_1543

These abundant cartographical imagery suggested the fascination of imagining how space extended far beyond a situated eye, and a sort of key to processing the extent of that dramatically expanded spatial expanse of the inhabited world.  But for Brouscon, as for le Testu, Java was both a continent of sorts, that extended to the pole, and needed to be accommodated by an extra flap of paper to be contained, but an uninhabited or at least unknown place in the “Terra Austral,” jutting up to Indonesia to reflect geographic tradition and, perhaps, to balance the landmasses distributed elsewhere on the chart:
Java in Brouscon's Map
The addition of far greater detail and qualitative content in the Vallard map developed the notion of the materiality of the map in the Dieppe school.  If we are struck most by its interior, the coastline of Java in the Vallard map suggest a detailed attentiveness to local toponymy, derived from Portuguese sources, of more specific scope, despite the lush detail of its interior, and a playful alteration of inks of different colors to add variety to its form:
java's Coastlines
The considerable local detail depicting something like the discovery of a pre-Adamic life is something of a counterpart to that expansion of the inhabited expanse in early world maps:   the clothes of its inhabitants, cast in somewhat neo-Orientalist garb as following their red-turbaned leader to greet arriving men, are paired with curious dwellings, customs, and styles of work, as well as a uniquely local bestiary and vegetation, as   well as different customs of social life: two figures on the left of this scene almost seem foreign observers, describing what they see:
Dancing Inhabitants.Vaillard

 

The narwhale rising from the waters just off the coast, shown without regard for the scale of islands or inlets on the coast, paired with a resting bobcat or lynx.  This image of the Eastern coast of Australia is only slightly embellished from Vallard’s original.

It’s been recently suggested that the inventive maps of the Dieppe school fabricated the entire continent out of geographic legends to evoke a potential land for colonization by the French monarch, as in this earlier 1566 Desliens world-map.  Indeed, these deluxe manuscripts reflect the broader interest of the materializing of wealth on the map, and, on the surface, seem to cast Java La Grande as something of the potential equivalent of what the Spaniards found in the Americas.

 

Nicolas_Desliens_ World Map 1566) with Java

 

The notion of this fabrication of continent seems absurd, but had confirmation in geographic theory.  Java La Grande is projected as a land of potential conquest and wealth, and is a survival of medieval written geographies that was transposed to a recognizably modern cartographical form, if it antedated the imagined expansive island of Taprobana, identified with Madagascar but often shown as a land of wealth, and in ways moved this target of European interest further East toward the Spice Islands and Indonesia.

The exquisitely tangible nature of the contents of the Vallard map may give some confirmation to its invention.  The upper register of this first image of the original Vallard atlas, now in the Huntington library, showed the region’s aboriginal inhabitants in a monochrome hues of striking similarity to a cave-painting as proud hunters bearing spears:

 

Dieppe atlas origianl

 

The image now held in Canberra is a striking copy of this image of the inhabitants of Java le Grande, which featured its  inhabitants in a procession across the newly mapped land:

 

horseback procession in Java on horseback

 

The placement of  initial folio perhaps as it was the most pressing communication of cartographical news–pays particular attention to the forms of habitation in Java, and the houses in which the inhabitants live and the palm-nuts on which they live:  as if to embody the information displayed in disembodied form in later world map projections:  in the years before Mercator’s Theatrum orbis terrarum, nautical charts concretized viewers’ material relation to spatial particularity.

 

First map Vallard atlas

 

The map offers something like a luxurious window into the newly discovered land for viewers to contemplate in ways that simple terrestrial projections did not allow.

The imagined continent of luxury and untold riches, filled with nutmeg and cloves as well as “idolatrous inhabitants,” made its way onto the globe by 1583, if somewhat assimilated to Antarctica:

 

Globe_terrestre_de_Jacques_Vau_de_Claye_(1583)

 

This might explain the staying power of nautical charts, based on observational practices of sailors and possessing a clearer pedigree as transcriptions of space, into the seventeenth century.  At the same time as material goods were arriving in Europe from west Africa and southeast Asia, maps provided something of a spatial catalogue to understand their arrival and place them somewhere in a lived topography, as much as they offered tools of orientation.  Java La Grande was attractive as something of an evidence of the inhabitation of a region later identified with Australia, before the arrival of Captain Cook, in ways that depicted the inhabitants that occupied its expanse in something of a romantic light.

 

Add. 7085, f. 1

4 Comments

Filed under Australia, Java La Grande, La Grande Jave, Nautical Charts, rutters