![]()
Part I. Contestations
1
On the epistemology of maps and mapping
De la Cosa, Mercator and the making of spatial imaginaries
Luis Lobo-Guerrero
For almost fourteen hundred years, European understanding of the earthly sphere and its relation to the universe (cosmographic knowledge) remained that described by Ptolemy in the second century AD. This was the world of Greco-Roman times, the oikoumene that covered a quarter of the earth’s surface (Snyder, 1997, 10–14). In Ptolemy’s Geographia, such a world began with the Canary Islands in the west and China in the east, the British Isles in the north and ‘Terra Incognita’ in the south. The oikoumene was a landmass partly surrounded by a series of disconnected waters, the Britannic Ocean to the north, the Indian Ocean to the south, the Occidental Ocean to the west, and a plethora of connected seas within the Mediterranean (Ptolemaeus, 1966 [1511]).
By the time of the Renaissance, the European conception of the world changed very quickly. It was a time of exploration and expansion in which Europe built on the experience of distant trade and travel resulting from the commercial revolution of the thirteenth century (De Roover, 1942; Lopez 1971). Distant trade implied long-distance sailing routes, and such routes required maps. Such was the context under which a particular tradition of mapping emerged in the Mediterranean and northern Europe; that of Portolan charts. These were maps intended for travel which also depicted commercial and political connections with uses in diplomacy and trade (see, e.g. Tobler, 1966; J. E. Kelley, 1977; Campbell, 2007). By the fifteenth century, when Iberian powers began to compete for discovery and conquest beyond the Ptolemaic world, first in the Atlantic, then in the Indian and the Pacific oceans, cartography began to depict different problems in the representation of space, a European depiction of spatiality that would quickly understand itself as global.
By 1500, the Majorcan Portolan practice, a cartographic tradition that corresponded to a mainly Mediterranean form of sailing, employed since Antiquity into the late Middle Ages, was in vogue. Pilots would either sail from port to port along a coast where rocks, shoals, and other obstructions, not always visible, posed perils to their ships, or would clear away from them by sailing out to sea ‘where courses were more direct, underwater hazards were less prevalent, and there was more room to manoeuvre’ (Ash, 2007, 510; Hewson, 1983, chaps 1–3). For longer voyages where coastal routes were impractical or impossible – for example, between Venice to Alexandria or from Norway to Iceland – pilots would seek to sail along stable courses by keeping a direction of travel and by estimating the distance covered (Ash, 2007, 510). This they could do with the help of a compass, by estimating their location in relation to the Polar Star and the sun’s zenith with basic instruments, and by resorting to some form of distance estimation system such as the chip log (see Waters, 1955; 1970). For Mediterranean sailing, where pilots were rarely out of visual contact with the coast for more than two days, Portolan maps offered useful rich coastal details and general political and economic information, as well as rhumb lines that facilitated the plotting of their ships’ courses (Campbell, 2007).
When Portuguese and then Castilian sailors ventured to Atlantic waters in the fifteenth century, first as a way to find routes towards the south of Africa and later west to ‘the Indies’, a different form of navigation started to be developed that would in turn lead to the emergence of a different cartographic tradition. In Atlantic waters pilots began to identify sea currents and winds which were different from those of the Mediterranean which made sailing through straight courses more difficult, if not impossible. They also ventured beyond the latitude and longitude of the Mediterranean and began to experience the effects of problems such as magnetic deviation (the difference between magnetic and true North) when trying to locate themselves. This demanded the adaptation of skills, ship design, sails, and navigation techniques and instruments. For such form of navigation, the straight sailing courses and coastal details offered by Portolan maps were of lesser value than information on sea currents, wind systems and, especially, ways of knowing latitude and longitude as a means for determining location.
This chapter tells a story of how the making of early modern cartographic representations at the transition between the Majorcan Portolan mapping tradition and the geometric projections that appear in the second half of the sixteenth century help reveal some of the most outstanding and useful elements of mapping for understanding the making of orders of governance in time. The chapter focuses on two maps, Juan de la Cosa’s 1500 ‘mappa mundi’, the first cartographic representation to show the Americas as a separate and single continent, and Gerardus Mercator’s world map of 1569, a projection that allows for a flat representation of a globe considered as the first modern geometrical depiction of global space. Both maps reveal a particular quest in the development of a European understanding of spatiality: the problem of location. They also reveal the usefulness of maps as instruments for travel and imperial business offering an opportunity to explore the particularities of how geopolitics were understood in the context of time. By focusing on issues such as how maps located waters and lands, how they represented latitude and longitude in their projections, how they plotted imaginary lines with political significance, how they represented empty and ‘silent’ spaces, how they presented and employed scales, how they integrated imagined and discovered lands into single cartographic charts, and how symbols, forms of writing, languages, and images constituted forms of rhetoric which help reveal specific forms of power (e.g. Harley, 2006; Tegeler, 2018; Boria and Rossetto, 2017) it is possible to reconstruct narratives that shed light on the making of a European ordering of global space; or, in other terms, of a making of European globality.
Rich as the problematisation of the making of spatial globality is, or better put, globalit(ies), if we were to move beyond the Eurocentric perspective, this chapter can only attempt to offer some general observations, contextualised through just a few references of a growing literature on what could be called the epistemologies of European-based modern cartography.1 In what follows, the chapter begins with an introduction of the two maps and the context, conditions and practices under which they were produced, continues with an analysis of the logics of representation involved in them, and concludes with some reflections on the epistemology of mapping for the understanding of the making of orders of the real.
De la Cosa’s mappa mundi
In 1500, Juan de la Cosa, owner and master of the Santa Maria, the largest of the three ships that travelled with Columbus on the first trip of ‘discovery’, produced a mappa mundi, a chart considered to be the oldest known European cartographic representation of the New World (Figure 1.1). For the first time, a mappa mundi displayed a unified continent in the Indies, what we today call the Americas. As an eyewitness who combined the roles of mariner, entrepreneur, and royal officer into the craft of mapmaking, de la Cosa’s representation of the world can be explored as revealing an imaginary of power embedded in the contexts of European imperial competition, the state of cosmographic knowledge and practices of the time, and a Christian account of colonisation and ordering of planetary space.
FIGURE 1.1 Mappa mundi, 1500, Juan de la Cosa; rotated to match modern map orientations.
Source: Original at the Museo Naval de Madrid (PD-1923).
The mappa mundi is an early example of a map that retains the Portolan detail on coasts and sailing courses but also begins to plot land according to estimated latitudes across the Atlantic. The map has a clear focus on location based on latitude, which is evident, for example, in the marking of the line of the equator and the Tropic of Cancer (the Northern Tropic) as reference lines. This is also evident when observing how the demarcated contours of land in the continent provide a general idea of the shape of the coastline and offers some latitudinal accuracy, to the degree that was possible at the time, around the Antilles region. Although the shape of the continent remains vaguely familiar to us today, as it advances north and south its location appears strange. This reveals the absence of a way of calculating longitude that affects how the north and south sections of the continent are depicted, almost as if they had been rotated ninety degrees with regard to the position they are ‘usually’ located in.
It is worth mentioning in passing that the tradition of establishing geographical location employing meridians and parallels is an application of Euclidean geometry and was already adopted by Ptolemy in the second century in his Geographia (Bagrow, 1945), although, as will be seen later, not with the geometric precision developed by Mercator in the late sixteenth century. When fifteenth- and early to mid-sixteenth-century Iberian cosmographers such as de la Cosa employed the use of parallels to indicate latitude in their maps, they did so based on their understanding of parallel lines in the Mediterranean which, in virtue of Ptolemy’s work, were considered equivalent in length. As a consequence, the projection of space offered by their maps was of a cylindrical form that slightly distorted the depiction of coastlines in the European seas and, as will be noted below, did not have much consequence for areas close to the equator. It did, however, have major implications for the depiction and the determination of location in regions beyond the tropics as evident in the mappa mundi. It is also important to note that the calculation of longitude at sea remained impossible until the invention of the marine chronometer in the eighteenth century, regardless of the offering of prizes since 1598 in Spain, for whoever could provide a solution to the problem of keeping the time of origin throughout a sea voyage.2 Other geographic phenomena of particular relevance for establishing location beyond the tropics, such as magnetic declination, although apparently known to the Chinese centuries earlier (Smith and Needham, 1967), and allegedly experienced by Columbus on his first trip (James E. Kelley, 1983), remained without a means of calculation in the West until the eighteenth century.
De la Cosa’s map was not intended for navigational use. It appears to have been commissioned by Archbishop Fonseca, then in charge of organising the trips to Indies, with the intention of presenting the global perspective of the discoveries to the Catholic Monarchs as well as to foreign dignitaries, possibly Italian (Martin-Meras, 2000, 80).3 Its most notable contribution was the depiction of the American continent as a landmass separated from Asia, a perspective Columbus died ignorant of in 1499. It will take until 1513 for the South Sea (Pacific) to be discovered by Nunez de Balboa confirming the separation of the continents, for the Pacific coasts of the Americas and for the coasts to start to be mapped.
Juan de la Cosa played a central role with vested experience and authority in the trips of discovery of the new world. After nine trips across the Atlantic, de la Cosa accumulated as much Atlantic sailing, navigational and cosmographical experience as was possible at the time. Partnered with the great names of Alonso de Ojeda and Americo Vespucci, Vasco Nuñez de Balboa and Rodrigo de Bastidas, he was considered by 1507 to be one of the five most experienced and knowledgeable sailors of Castille. In such capacity he was summoned by King Ferdinand to the Junta de Navegantes, a technical commission tasked with exploring the possibilities of opening a western route through which to reach the islands of the species, the ‘Especieria’. Although the task was eventually only achieved in 1519 by Magallanes and Elcano with their circumnavigation of the globe (Arranz Márquez, 1982, 87), de la Cosa’s experience had more immediate effects. In 1503 the Casa de Contratación de Sevilla was established as what might be understood today as a super-ministry in charge of most American affairs, and by 1507, two years before de la Cosa’s death, the position of ‘piloto mayor’ was created. This new role was tasked with maintaining an up-to-date chart of the world, the Padrón Real, which included information derived from reports and corrections provided by pilots returning from the Indies. The basis for such a map would be de la Cosa’s mappa mundi.
De la Cosa’s map also played a role in depicting the Spanish geopolitical claim within the Treaty of Tordesillas. This was a treaty whereby Castile and Portugal agreed to divide the world in two parts based on the 1493 Pope Alexander VI’s Bull Inter Coetera, an instrument which granted the rights to discovery and colonisation to Spain west of an established meridian and to the Portuguese to the east. The dividing line was established as being 370 leagues west of Cape Verde islands off the coast of Africa (Goodman, 2002, 53M CS). Determining the exact location of the line was, however, contentious, given the difficulties of demarcating a line at sea and of locating it on a map in the absence of precise geodesic measurements. After the Magallanes-Elcano’s circumnavigation of 1519, during which they reached the Moluccas islands and the Philippines, the location of the line became of paramount importance since the ‘anti-meridian’ was to determine the right of access to trade in spices in such longitudes, and also, with China. Whereas it was not disputed that Portugal reigned in the Indian Ocean, it was not clear where the limit of such territory was in Asian waters. In 1524 Charles V convened a conference in Badajoz including jurists, cosmographers and commissioners from both sides to settle the matter (Portuondo, 2013, 66). Whereas there is no agreement among experts on what the line at the very east of the American landmass in de la Cosa’s map represents, it is taken by some to be the Spanish representation of the Tordesillas line, a strong visual political statement to competing European powers and a depiction of the territorial vastness of the nascent empire.
Mercator’s world map
Sixty-nine years after de la Cosa’s mappa mundi, Gerardus Mercator produced in Flanders a world map (Figure 1.2), a new cartographic representation using a geometric projection that represented sailing courses of constant bearing as straight lines in the style used by Ptolemy in his Geographia but with technical corrections. As noted by Snyder, Ptolemy’s technique in mapping the oikoumene, his ‘scheme of describing the position of places by their longitude and latitude, could theoretically be applied to the whole sphere … it was a geographical idea of elegant simplicity’ (Snyder, 2007, 370).
FIGURE 1.2 Gerardus Mercator’s world map, Nova et Aucta Orbis Terrae Descriptio ad Usum Navigantium Emendate Accommodata, 1569.
The map appeared divided into eighteen sheets providing a flat representation of the world, as a cylindrical projection, but this time, spacing the parallels more widely as the poles are approached to prevent the deformation offered in previous maps (Snyder, 1997, 45). His contribution aimed to correct the distortions that resulted from conical projections of space derived from the Ptolemaic assumption on the length of parallels, such as that implicit in de la Cosa’s map. Explicitly intended for navigation, as noted by Mercator in the description contained in the map itself, it sought to offer practical resources with which to find a ship’s location at sea and the planning of voyages based on the calculation of distances. It showed all lines of constant bearing (loxodromes or rhumb lines) straight, including parallels and meridians. Mariners ‘could then follow a single compass setting (adjusted for magnetic declination) based on the bearing or azimuth of the straight line connecting the point of departure and destination on the map’ (Snyder, 1997, 45).
However, the map was, as is technically called, a ‘conformal map projection’ derived from a stereographic one, and as such did not allow for observing distances correctly (Snyder, 1997, 45). For its use in navigation the mariner required mathematical training to handle the complex calculations to reveal actual distances. This complication explains why it took several decades to be of practical use. In 1594 a set of tables produced by Edward Wright for calculating the length of the lines were published (Blundeville, 1594), easing the use of the projection in navigation and making the map of use for ordinary mariners.
Although aimed as an aid for navigation, Mercator’s projection was soon adopted for geographical purposes within the context of European imperial competition. In 1587, Romuldus Mercator, son of Gerardus, published a ‘Two Hemisphere World Map’ employing his father’s projection. By 1595, Jodocus Hondius, a Flemish cartographer and religious refugee in London, adapted Romuldus’s map by shifting the central meridians by ninety degrees so that Europe, Africa and the Ameri...
