Famous charts and forgotten fragments: exploring correlations in early Portuguese nautical cartography

ABSTRACT

The authors of the well-known collection Portugaliae Monumenta Cartographica hinted at connections between two anonymous portolan charts from the beginning of the sixteenth century, namely the portolan chart at the Bibliothèque Municipale of Dijon and a fragment of a chart kept in Lisbon in the Archive at Torre do Tombo. Later, they also mentioned affinities between those two charts and the famous chart known as Kunstmann III. However, they did not pursue these observations further. The present paper proceeds from where those researchers stopped investigating and proposes a fresh look on this cartographic material by combining a traditional historical approach with modern digital techniques. First, a comparative study of the toponomy of a common area of the charts will be presented. Later, each chart will be examined with the help of cartometric methods to access their implicit geometry. The advancements on the study of correlations between these charts will be shown, thus confirming that the combination of traditional and digital methods of investigation open very promising perspectives to the study of unsolved questions in the History of Cartography.

RÉSUMÉ

Les auteurs de la très célèbre collection Portugaliae Monumenta Cartographica ont suggéré des similitudes entre deux cartes portulans anonymes du début du seizième siècle, à savoir la carte portulan la Bibliothèque Municipale de Dijon et un fragment de carte conservé à Lisbonne, aux Archives Nationales Torre do Tombo. Plus tard, ils ont également mentionné des affinités entre ces deux cartes et la fameuse carte de Kunstmann III. Pourtant, ils n'ont pas approfondi ces observations. Ce papier commence à l'endroit où ces chercheurs ont terminé leur recherche et propose un regard neuf sur ce matériau cartographique en combinant l'approche historique traditionnelle et les techniques numériques modernes. En premier nous présentons une étude comparative des toponymes sur une zone géographique commune aux différentes cartes. Puis, chaque carte sera étudiée à l'aide de méthodes cartométriques pour connaitre leur géométrie implicite. Enfin nous montrerons les avancées concernant les corrélations entre ces cartes, confirmant ainsi que la combinaison de méthodes recherches traditionnelles et numériques ouvre des perspectives très prometteuses pour l'étude de questions non résolues dans l'histoire de la cartographie.

KEYWORDS:

• Cartometric analysis
• portolan charts
• Portuguese cartography
• 16th century

Acknowledgements

I would like to thank Joaquim Alves Gaspar, Gregory McIntosh and Šima Krtalić for key suggestions and proofreading, and to Gonçalo Dias for the assistance with the error analysis.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributor

Historian of Science. PhD thesis: ‘The influence of Pedro Nunes in navigation in the 16th and 17th centuries: a study in transmission of knowledge.’. Currently working in the Project MEDEA-CHART, hosted by the Department of History and Philosophy of Sciences, Faculty of Sciences, University of Lisbon and the Interuniversity Center for the History of Sciences and Technology (CIUHCT), [https://www.medea-chart.org].

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

1 Dijon, Bibliothèque Municipale de Dijon, Ms. 550.

2 Lisbon, ANTT, Fragmentos, cx. 20, n.° 7.

3 The chart is known as Kunstmann III, or Kunstmann n° 3, due to its sequence in the catalogue (Kunstmann, von Spruner, & Thomas, 1859). The original chart was lost during World War II. In 1836, a hand-drawn copy of the complete chart in colour was made by the German artist Otto Progel. This copy was donated to the Bibliothèque Nationale de France (Paris) in 1843. See: Département Cartes et Plans, CPL GE B-1120 (RES). There is also a photographic reproduction of the original K3 chart, published in four sheets in (Stevenson, 1903, no. 3) (artotype, four sheets, 1220×830 mm).

4 The use of cartometric methods to investigate the geometry of the medieval and early modern cartographic material is one of the goals of the MEDEA-CHART, hosted by the Department of History and Philosophy of Sciences, Faculty of Sciences, University of Lisbon and the Interuniversity Center for the History of Sciences and Technology (CIUHCT). See: [https://www.medea-chart.org].

5 MapAnalyst is a freeware computer application developed by Bernhard Jenny and Adrian Weber. See (Jenny & Weber, 2005), Free access online at: http://mapanalyst.org.

6 (Santarém, 1919, pp. 129–131).

7 In fact, this does not completely exclude other origins, because it was common among cartographers to use decorative elements from different sources. Alfredo Pinheiro Marques wrote: “As for the proximity of the decorative elements to those of the Catalan/Majorcan models, we know that this was common in the earliest Portuguese charts.” (Marques, 1987, p. 84).

8 A significant example was his opinion that the chart had to be made before 1429 because of the presence of the papal flag in the city of Avignon. Yet, later charts, such as the portolan by Jorge de Aguiar, made in 1492, still portray the same flag.

9 See for example (Winter, 1950, p. 39).

10 The ink in the Dijon chart is very faded in that area, nevertheless, by comparing it with the place names in the K3, it is possible to discern that Mazalquivir and Oran are both in red. More about the examination of the toponomy will be presented below.

11 Technical examination and advanced imaging techniques might provide more information about the original form of the drawing.

12 The Book of the Knowledge of all the Kingdoms, Lands and Lordships that are in the World (Castilian original title: Libro del conosçimiento de todos los reynos y tierras y señoríos que son por el mundo), is a fourteenth-century (c. 1385) anonymous armorial manual. Four extant manuscript copies are known. The edition used as a reference in the present study is the English edition (Markham & Jiménez de la Espada, 1912).

13 See, for example, the chart Chart of the Mediterranean, Black Sea and Western Europe attributed to Angelino Dulceti, produced between 1339 and 1350, London, British Library, Add. 25691; the chart of the Mediterranean and Black Sea by Petrus Rosselli, produced between 1450 and 1475, New Haven, Beinecke Rare Book and Manuscript Library, Art Storage 49cea 1425; the anonymous chart of the Mediterranean, Black Sea and western coast of Europe, produced between 1450 and 1500, San Marino, Huntington Library, HM 1548.

14 For example, the anonymous chart (attributed to Pedro Reinel) at the Bayerrische Staatsbibliotheck, Cod. Icon. 140 (chart n°82), (Cortesão & Teixeira da Mota, 1960, Vol. I, pp. 23-24). Other charts present Spanish flags in the same locations in North Africa, for example, a chart by Salvator de Pilestrina (made in 1533) kept in Toledo, Biblioteca de Castilla-La Mancha, Ms. 530.

15 Although not improbable, the possibility that an eventual cross in Tripoli could have faded and disappeared along the years is remote. This is not evident with a visual inspection and could eventually be confirmed with the use of non-invasive physical or chemical techniques.

16 If the marking near Monte Sines is not a cross - a highly unlikely scenario - the dating of the crosses may even need to be shifted. In this alternative hypothesis, they could have been added at a time when the only Spanish possessions in the area were at Peñón de Velez and Oran. Hence, in such a scenario the drawings and crosses could have been added after 1564, the year when the Peñon de Velez was recovered after being lost arround 1522.

17 (Cortesão & Teixeira da Mota, 1960, Vol. I, p. 5). Another reference can be found in (Cortesão, 1971, p. 218).

18 The restoration of Sardinia to the House of Aragon occurred in 1493 and re-established the Kingdom of Naples.

19 Discussion about these dates of production can be found in (McIntosh & Gaspar, manuscript in preparation).

20 There is also a photographic reproduction in (Cortesão & Teixeira da Mota, 1960, Vol. I, Plate 6).

21 “There are some affinities with the so-called Kunstmann III chart of c. 1506 (Vol. I, pp. 15-6, Plate 6), which might suggest a common authorship.” (Cortesão & Teixeira da Mota, 1960, Vol. V, p. 187).

22 The Cantino Planisphere, was chosen for practical reasons. This planisphere is Portuguese-made, produced about the same period as the other three charts and depicts the coastline under review. It is probably the most relevant product of early Portuguese cartography and was used as a source of information for subsequent charts. It displays the world known at the time and was drawn in three parchment leaves measuring overall 105×220 cm. Although it is a planisphere, the area of the Mediterranean, used data from portolan sources.

23 Paul Gaffarel made a list of place names for the Dijon chart, see (Gaffarel, 1874, pp. 168–199). The list is useful, but it includes some mistakes. (McIntosh & Gaspar, 2018) presents a survey of the toponomy of Brazil, Africa, Newfoundland, Baltic and Scandinavia. As far as I know, nothing has been published before now about the toponomy of the ANTT fragment.

24 Progel’s coloured copy of the K3 was used to determine red ink place names.

25 This procedure was advanced in (Cortés, 1551), fl. lxiiii r: “y primeramente se han de escribir de colorado los puertos y cabos principales, y famosas ciudades, y otras cosas notables: y todo lo demás de negro. (…)”. Translation: “and first write in colour the principal ports and capes, and other notable features: and everything else in black”, please note: the translation is mine. (McIntosh, 2013, pp. 27–28), supports this procedure. Nevertheless, Ramón Pujades is more inclined to the opposite procedure, that is, he states that black names were writen first, (Pujades i Bataller, 2007, pp. 480–481). For example, the already mentioned anonymous chart at Bayerrische Staatsbibliotheck, Cod. Icon. 140 (chart n°82), only shows black place names and the spaces for the red place names are empty.

26 See (Campbell, 2014).

27 If the date of production of the K3, as suggested in (McIntosh & Gaspar, 2018), i.e. 1501-1506, is to be accepted, than the K3 must count as the first chart to have both Villa Franca and Porto Pisano in red.

28 The internal geometry of portolan charts is associated with the representation of the actual geographic features (coastlines, islands, etc.) and reflect the navigational techniques used to acquire the geographical information. See (Gaspar & Leitão, 2018).

29 For examples of analogous grids from charts of the period see (Gaspar, 2010), (Gaspar, 2012), (Dias, 2018). Recently, a systematic survey of the geometry of early charts has been conducted within the MEDEA-CHART project. More results of this investigation are expected to be published in the near future.

30 For more on the influence of magnetic declination in the geometry of portolan charts see (Gaspar, 2010) and (Gaspar, 2008).

31 In the MapAnalyst application, the Dijon chart was used as the “new map” and the ANTT fragment and K3 were used as the “old map”.

Additional information

Funding

Project MEDEA-CHART has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement n° 714033-MEDEA-CHART/ERC-2016-STG). H2020 European Research Council.