https://orcid.org/0000-0001-8304-4583
![Sample our Geography journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5937/TOC-Subject-Sample-2021-Geography.jpg"})
ABSTRACT
A chart of the sixteenth century is extant, depicting the southern hemisphere and containing the earliest known representation of the southeast coast of South America, in the wake of Magellan and Elcano's circum navigation. In this paper, it is argued that astronomical observations of longitude were accommodated in the representation, and that the chart was produced in the specific context of the Juntas of Badajoz-Elvas, held between the representatives of the Spanish and Portuguese Crowns, to discuss the location and possession of the Spice Islands. It is further shown that the chart was produced using the information brought to Seville by the ship San Antonio, which had deserted the fleet in October 1520, before the passage to the Pacific Ocean was found. It is concluded that this chart presents a unique historical milestone in the history of cartography, containing the earliest material evidence of the effective use of astronomical methods to determine longitude in a nautical context.
Acknowledgements
I warmly thank Romeu Gaspar for his support in assessing the errors associated with astronomical determinations of longitude in the sixteenth century, which has permitted more solid conclusions to be drawn about how this chart was constructed. This work was made possible by the collaboration of Šima Krtalić in the in situ examination of the manuscript and the research that followed.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 The designation chart, instead of map, will be used throughout this text notwithstanding the fact that this is not strictly a nautical chart and was not intended for navigation. Despite of the use of a map projection foreign to marine navigation, the representation was copied from traditional charts, and the cartographic conventions in the depiction of the coastlines and names are those of nautical cartography.
2 The anonymous chart known as Kunstmann IV planisphere (c. 1519), attributed to Jorge Reinel and Pedro Reinel, was lost during World War II. A black and white photograph is extant, as well as a coloured facsimile drawn in 1843 by Otto Progel, now kept at the Biblothèque nationale de France (CPL GE AA-564 (RES)). The Miller Atlas (c. 1519) is a luxurious manuscript containing ten charts and a map of the world, whose cover page is signed and dated by the Portuguese cartographer Lopo Homem. It is considered to have been also drawn by Pedro and Jorge Reinel and is now kept at the Biblothèque nationale de France (Res. Ge. DD. 683). See Gaspar and Krtalić (Citation2023: 127–135; 136–142).
3 The anonymous chart known as Castiglione planisphere was completed in 1525, in the Casa de la Contratación of Seville, to be offered to Count Baldassare Castiglione, apostolic nuncio to Pope Clement VII in Madrid, on the occasion of the marriage of Emperor Charles V (Carlos I of Spain) with Princess Isabel of Portugal. It is now kept in the Biblioteca Estense Universitaria, Modena (C.G.A.12). See Cortesão and Mota (Citation1987: 95–97); Martín-Merás (Citation1993: 91–99); Martínez (Citation1994: 183–192) and Gaspar and Krtalić (Citation2023: 168–172).
4 An explanation is needed about why that course of 209° does not coincide with that measured on the projection of this modern map. The reason is that loxodromic directions are not conserved on a cylindrical equidistant projection (the projection of the modern map), a property only observed on the Mercator projection. But nautical charts of the time, where the parallels were equally spaced, were not constructed using an explicit map projection. Directions measured with the marine compass were transferred directly to the chart, not correcting for magnetic declination, and ignoring that they had been measured on the spherical surface of the Earth. The result led to geometric inconsistencies, in the sense that only the specific tracks used to construct the charts were correctly represented. All other directions measured on them were affected by variable errors. This fact was first noticed by the mathematician Pedro Nunes (Citation2002) in his Tratado en defensam da carta de marear of 1537 (‘Treatise in defense of the navigational chart’), where he drew attention to the fact that most directions, including the north-south direction, were incorrectly depicted on the contemporaneous charts. The significant detail to note in this specific situation is that the coast of Brazil on sixteenth-century cartography would be correctly oriented, for navigational purposes, if it were represented on the charts in accordance with the compass courses made by the ships along the shoreline. The effect of a spatially variable magnetic declination on the orientation of the coastlines would not be noticed by the mariners because navigation when also made with the same type of magnetic compasses. A more detailed explanation of these subject is in Gaspar (Citation2007, Citation2010: 66–73).
5 The fact that the fleet did not navigate on a straight line between Rio de la Plata is irrelevant here. Using a composite track more representative of the real route along the coast would produce a similar result.
6 For the list of astronomical observations of San Martín, see João de Barros (Citation1628: Capitulo X, fl. 147). For the value of longitude determined by San Martín (61° measured from Seville), see Fernão Lopes de Castanheda (Citation1924: 159–160). The principle underlying the determination of the longitude at the time, which had been known since antiquity, consisted in comparing the local times of the occurrence of certain astronomical phenomena (like an eclipse) in two different places. Knowing the longitude of one of them, the difference between the two local times would immediately give the longitude of the other, considering that the celestial sphere apparently rotates 15 degrees per hour (or 1 degree every 4 minutes). Because of the inconvenience of having two observers in different places, whose measurements could only be compared a posteriori, astronomical ephemerides were available giving the local times of astronomical phenomena in reference places, like conjunctions and oppositions of heavenly bodies. For an estimation of the errors associated with the astronomical determination of longitudes during the sixteenth century, see Steele and Stephenson (Citation1998) and Gaposchkin and Haramundanis (Citation2007).
7 A lengthier explanation of the various errors affecting the observation is provided by Romeu Gaspar (Citation2023).This author suggests that San Martín purposedly discarded the Toledo tables, either because he detected a typographic error in the position of the Moon (which explains the different local times of the phenomenon), or because the longitude of San Julián determined with them would be inconsistent with the navigational data collected along the east coast of South America.
8 For a more detailed description of the depiction of Southeast Asia on this chart, see Gaspar and Krtalić (Citation2023).
9 This letter was sent from Badajoz, on 18th May 1524, by three representatives of King João III of Portugal: Francisco de Mello, Pedro Afonso de Aguiar and Diogo Lopez. See ‘Carta dos deputados que tinham sido enviados para tratar do negócio de Maluco’. Badajoz, 18th May 1524, in As Gavetas da Torre do Tombo, IV (Gav. XV, Maços 1–15). Lisboa, Centro de Estudos Históricos Ultramarinos, 1964, pp. 341–348.
10 According to the terms of the Treaty of Tordesillas, the Line of Demarcation between the two Iberian hemispheres should pass 370 leagues (a little more than 2,000 km) to the west of the Cape Verde islands. Although no explicit agreement was reached on this matter, the tacit interpretation was that the westmost island of the archipelago should be taken as origin of the counting.
11 Before Magellan’s mission, all extant charts of Portuguese origin depicted the Moluccas in the Spanish hemisphere. That is the case of the anonymous chart of the Indian Ocean (c. 1517), attributed to Pedro Reinel, and of the Kunstmann IV planisphere, made by Jorge and Pedro Reinel c. 1519, to be offered to King Carlos I of Spain. See Gaspar and Krtalić (Citation2023: 136–142).
12 In fact, the correct longitudinal position of Bacan relative to the antimeridian is about five degrees further to the west, that is, well within the Portuguese hemisphere.
13 That these longitude measurements were related to the ongoing discussion about the position of the Spice Islands, which were situated within the Spanish hemisphere on contemporary Portuguese cartography, seems obvious. The cause for the mistake is here implicitly attributed to an exaggeration of the Indian Ocean width, caused by the incompetence of pilots. However, both claims were wrong: the width of the Indian Ocean was, in fact, slightly underestimated on the charts; and the real cause for the displacement of all lands to the east of Africa was its eastward displacement and stretching, caused by the effect of magnetic declination on the compass directions used to construct the charts. This fact was fully explained by João de Castro (Citation1538), in his Roteiro de Lisboa a Goa (Rutter from Lisbon to Goa).
14 Although the observation of eclipses was universally acknowledged as the best method to determine longitude, the astronomers of the time were aware of the fact that only repeated observations taken over a long period could guarantee accurate results. The earliest extant Portuguese charts depicting the region after 1524 are those contained in the anonymous atlases of c. 1537 attributed to Gaspar Viegas, where the Moluccas are depicted with a longitudinal error of less than one degree. By that time, the Portuguese have already had the opportunity to make several astronomical determinations of longitude in the region. For a list of the position of the Moluccas on European charts of the sixteenth century, see Gaspar and Krtalić (Citation2023, 233–243).
15 How the manuscript was later found in possession of the Ottoman Turks is more difficult to explain. It might well be the case that it was taken from Lisbon to Spain, together with other documentation that included the notes of Andrés de San Martín, during the Iberian Union (1580–1640). This large European territory then ruled by the Spanish monarchs may make it easier to explain how it ended up in Turkish hands.
Additional information
Funding
Notes on contributors
Joaquim Alves Gaspar
Joaquim Alves Gaspar is a retired officer of the Portuguese Navy, specialist in Navigation, with a Master's degree in Physical Oceanography and a Doctorate in Geographical Information Systems. His research activity has been focused on the history of medieval and early modern nautical cartography, using numerical methods of cartometric analysis and modelling. He is now the Principal Investigator of the project Medea-Chart, funded by the European Research Council and hosted by the Faculty of Sciences, University of Lisbon, Portugal (https://www.medea-chart.org/about).