Title: De navigatione

Author: Benedetto Cotrugli

Date: 1465

Description: Benedetto Cotrugli’s De navigatione, composed in 1464–1465, is the earliest known European manual of navigation. Two manuscripts of the work survive; in both a space is left blank for a mappamundi alluded to in the text. In his article for Imago Mundi (65:1 2013, pp. 1-14) Chet van Duzer identifies three mappaemundi, in different works, as having come from or been derived from Cotrugli’s. The first is the unstudied mappamundi that was added to the beginning (fol. 2v) of Biblioteca Apostolica Vaticana, MS Ottob. lat. 1417. The second is the only mappamundi in any of the manuscripts of Cristoforo Buondelmonti’s Liber insularum archipelagi (Staatsbibliothek zu Berlin–Preussischer Kulturbesitz, MS Hamilton 108, fol. 81r). The third is the unstudied mappamundi in the Biblioteca Medicea Laurenziana, MS Ashburnham 1077, fol. 4v. The last two mappaemundi are similar to that in MS. Ottob. lat. 1417, and there are reasons to believe that they were copied from mappaemundi in manuscripts of Cotrugli’s work. In the following monograph, the toponyms on the Vatican and Berlin maps are compared, and the implications of these mappaemundi for the format of Cotrugli’s autograph manuscript and the diffusion of his work are explored.

Benedikt Kotruljevic (1416–1469), better known by his Italian name, Benedetto Cotrugli, was born in Ragusa (Dubrovnik, Croatia) and became a powerful merchant and diplomat whose influence was felt in southern Italy, Sicily and North Africa. Cotrugli had an excellent humanist education and was an innovative thinker, as is demonstrated by his writings. Cotrugli’s De navigatione is divided into four books. The first begins by describing the nature of water and the different oceans and seas. It has a long chapter on the qualities of ports, and ends with a brief geographical description of the whole world, along with a list of all the provinces in Europe, Africa and Asia. Book Two opens with a chapter on navigation in antiquity and then explains the various types of ships and their officers. In Book Three, Cotrugli discusses winds and meteorological phenomena, the signs of the zodiac, the moon and the planets, time, and seasickness. Book Four begins with praise of Ptolemy’s Geography. This is followed by a chapter about nautical charts, and then a portolan (written description) of the different ports, islands, rivers and other landmarks of the Mediterranean, with the directions and distances from one to another.

Neither of the two surviving manuscripts is the original, both having been made some years after Cotrugli wrote. The first manuscript is in Yale University’s Beinecke Rare Book and Manuscript Library, MS 557.4 There is no evidence that allows a precise dating of this manuscript, but it is probably from the last quarter of the 15th century and thus made just a few decades after Cotrugli completed his book in 1464–1465. The Yale MS is small, 20.5 x 14 cm (8 x 5.5 in.). The text abruptly ends in the portolan in Book Four, and is thus incomplete. Two of the four extant illustrations, a diagram of the elemental spheres on folio 7r and a zonal mappamundi on folio 10v, are also unfinished. Book One ends with the above-mentioned description of the world and list of provinces, and on fol. 23r is the rubric De Scriptio Totius Orbis—a phrase that could apply to a world map, as well as to a verbal description of the world—and a blank space. The intended world map is not, however, present.

The second manuscript is in the Lawrence J. Schoenberg Collection, MS LJS 473.6 at the University of Pennsylvania. This manuscript probably dates from the end of the 15th century, and it too is small, 21 x 13.5 cm (8.25 x 5.3 in.). The text has been transcribed, translated into modern Italian, and published. In this manuscript the diagram of the elemental spheres and the zonal mappamundi (folios 8r and 12r respectively) are complete and are colorfully painted. At the center of the diagram of the spheres, the earth is represented by a small and geographically vague mappamundi. On folio 27r, near the end of Book One and after the list of provinces, the rubric Descriptio Totius Orbis is written with blank space left for a mappamundi, but again no map is found.

Diagram of the elemental spheres with a small mappamundi representing the earth in Cotrugli’s De navigatione. The visual style of the mappamundi is similar to that of the exemplar shown below, which helps confirm the ascroption of that map to Cotrugli.

Florida, Lawrence J. Schoenberg Collection, MS LJS 473, fol. 8r

In a work about navigation, particularly one describing the use of nautical charts, one might expect to actually find charts. One wonders, thus, whether in Cotrugli’s original manuscript, the portolan relating to a voyage around the Mediterranean was followed by an atlas of nautical charts, perhaps similar to the contemporary atlases by Grazioso Benincasa. The fact that a work on navigation would include a mappamundi that, although useful for illustrating a general description of the world, is useless for navigation, is typical of the mixture frequently found in the 15th century of these cartographical types. Andrea Bianco’s atlas of 1436 (#241), for instance, consists of nautical charts, a mappamundi and a Ptolemaic world map without any indication that one is to be preferred to the other. A manuscript in the Vatican made before 1438 opens with Guillaume Fillastre’s introduction to Pomponius Mela, which is followed by the text of Mela’s De situ orbis. Between them, on folio 8, is a Ptolemaic world map painted by Pirrus de Noha. Noha’s map (#239) is geographically quite at odds with its manuscript context; in particular, it shows both the Indian Ocean and the Caspian Sea as surrounded by land, while Mela maintains the opposite. The Catalan Estense mappamundi of 1460 (#246) presents nautical-chart data in the format of a circular mappamundi. The tendency to mix cartographic genres in the 15th century will be important in the following discussion.

Context of Cotrugli’s Missing Mappamundi

Cotrugli’s brief geographical description of the world precedes the areas left blank for the mappamundi in both manuscripts. The text is an adaptation in Italian of part of Ptolemy’s Geography, Book 7, Chapter 5, which is titled “A Descriptive Summary of the Map of the Inhabited Earth”. On the one hand, Cotrugli borrowed a passage that addresses exactly the subject that interests him. On the other hand, according to Van Duzer the decision to introduce a mappamundi with an excerpt from Ptolemy in a book about navigation is bold, even in the 15th century context of the ready mixing of cartographical types. Here is Cotrugli’s description in translation.

We shall give a description of the habitable part of the world, which is the goal of our navigation. This habitable part ends in the terra incognita where the sun rises, which lies among the eastern peoples of Great Asia, namely the Chinese and the Gerunna*, and so Asia is said to be the first part of the world because it is near the sunrise. In the South there is likewise terra incognita that surrounds the Indian Ocean and also embraces southern Ethiopia, the region called Agisimba. There is also terra incognita in the West that surrounds the Gulf of Africa by Ethiopia and the western ocean, which lies in the furthest West. In the North there is the ocean that unbrokenly surrounds the British Isles and the most northerly parts of Europe, and this is the ocean they call the Deucalidonian and Sarmatic. Moreover, the terra incognita has other boundaries, specifically that in the far north of Asia, namely Sarmatia, Scythia and Serica. Then there are the seas that are surrounded by land: our Tyrrhenian Sea with its various adjoining bays, and those of the Adriatic, the Aegean, the Propontis, Pontus [Black Sea], and the Meotian Swamp [Sea of Azov]: [their waters] flow out into the Ocean through the Sea of Hercules or Strait of Gibraltar, much as at Chersonessus [Strait of Kerch]. This narrow stretch of sea surrounds the Hyrcanian or Caspian isthmus, creating from the land the similitude of an island*, and they say the same of the Indian Ocean, which together with all its gulfs—the Arabian, Persian, Gangetic, and that which they call the Great Gulf—is encircled by land on every side. Here, [of] the three largest parts [of the world,] Asia joins Africa just north of Arabia, which land separates our sea from the Arabian Gulf, and they are also joined by the terra incognita that surrounds the Indian Ocean. Asia joins Europe in the north, via the land between the Meotian Swamp and the Sarmatic Ocean, just where the Tanais River [the Don] is. Africa is separated from Europe only by the Atlantic Ocean, and the two continents do not touch anywhere.

*I have not been able to identify the Gerunna; Ptolemy speaks rather of the Seres in the passage that Cotrugli is translating here (Ptolemy, Geography, 7.5). Cotrugli’s translation of Ptolemy 7.5 is rather rough here: what Ptolemy says is that the Caspian Sea is surrounded on all sides by land and has the shape of an island.

This description of the world is followed by the list of provinces. Ninety-four provinces in all are named, under the relevant headings: in Europe, thirty-four; in Africa, twelve; and in Asia, forty-eight. On a number of medieval mappaemundi lists of provinces occupy the three parts of the world, and indeed maps of this type have been baptised ‘list maps’. Evelyn Edson, Mapping Time and Space: How Medieval Mapmakers Viewed Their World characterizes the list map as ‘closely related to the T-O . . . the basic cartographic frame, usually a T-O, is used as the space for a list of places in each region, most often a list of provinces’. In some English mappaemundi the provinces derive from the Historia Brittonum traditionally ascribed to the 9th century Welsh monk Nennius. These lists of provinces are much shorter than Cotrugli’s, however, with fifteen provinces in Asia, twelve in Africa and fourteen in Europe. Hence Cotrugli, while following the tradition of associating lists of provinces with mappaemundi, used a new source, extracted from Book 8, Chapter 2 of Ptolemy’s Geography, thus boldly mixing cartographic types.

The Mappamundi in BAV MS Ottob. lat. 1417

Although Cotrugli’s mappamundi was never drawn or painted into the Yale or the Schoenberg manuscript of his De navigatione, an example of it does survive in an unrelated manuscript, namely Vatican City, Biblioteca Apostolica Vaticana, MS Ottob. lat. 1417, an illustrated copy of Justin’s Historiarum Philippicarum libri XLIV. Justin, a Roman historian, is thought to have composed his work between the second and fourth centuries AD; the Vatican Library manuscript is dated 17 June, 1460. The manuscript has been described several times, but the mappamundi on folio 2v is generally unknown. It is not in Destombes’ catalogue of medieval mappaemundi, does not feature in other literature on medieval maps, and is not discussed in detail by any of the scholars who describe the manuscript.

Some manuscripts of Justin’s work are titled Historiae Philippiace et totius mundi origines et terrae situs, that is, The Philippic Histories and the Origins of the Whole World and the Description of the Earth. This title, together with the geographical digressions in the text, might suggest that it would be appropriate to illustrate the work with a map, but there is no tradition of a map integral to manuscripts of Justin’s treatise. Ottob. lat. 1417 is the only exemplar to contain a map. This map, however, was not originally part of the manuscript. Somehow Franz Rühl (Die Textesquellen des Justinus) is the only scholar who has noticed that the first two folios of Ottob. lat. 1417 (which include the map) were added to the manuscript after its creation, a conclusion confirmed by differences in spelling and handwriting between map and text. For example, compare the handwriting of Macedonia in lower right of the map with Macedonibus on folio 28v, and the spelling of Asia on the map with Asya in the text (fols 5v and 20r), and Scythia on the map (left side) with Scytis in the text (fol. 5v). Illustrations of the map and of these folios are available in Buonocore, Vedere i classici, 415–17.

According to Van Duzer, the mappamundi now in Ottob. lat. 1417 came from a manuscript of Cotrugli’s De navigatione. This is confirmed first, the general style of the map in Ottob. lat. 1417 is similar to that of the small mappamundi in the diagram of the elemental spheres on folio 8r of the Schoenberg manuscript of Cotrugli’s De navigatione (see above). Although the geographical details of the Schoenberg map are largely unidentifiable, like the larger map in Ottob. lat. 1417, it depicts the Earth with deeply indented coastlines and gives graphic emphasis to mountain ranges. The stylistic similarity between the small world map in the Schoenberg manuscript, which is part of Cotrugli’s work, and the large one in Ottob. lat. 1417 renders plausible the suggestion that the large mappamundi comes from a manuscript of Cotrugli’s De navigatione.

Much stronger evidence for the connection between the mappamundi in Ottob. lat. 1417 and Cotrugli’s work is found in the close correspondence of the provinces named in the margins of the Vatican mappamundi and those in Cotrugli’s De navigatione. The headings of the lists on the map indicate forty-eight provinces in Asia, twelve in Africa, and thirty-four in Europe, just as Cotrugli does. The first five provinces in Asia in Cotrugli’s text are Pontus, Bithynia quae proprie Asia dicitur, Phrygia Magna, Lycia, Galatia, and the first five provinces in the list on the map are, aside from minor spelling differ- ences, exactly the same: Pontus, Bithinia que proprie Asya dicitur, Phrigia Magna, Licia, Galatia. There are a few differences between Cotrugli’s names and those on the map in Ottob. lat. 1417—for example Tyle insula appears in the list for Europe in Cotrugli’s text but is not mentioned in the list on map—but the similarities are simply overwhelming. Moreover, the provinces named in the list are not only the same as Cotrugli’s, but are also presented in identical order (which is different from that of Cotrugli’s source, Ptolemy’s Geography, 8.2).

Unexpected things happen in history, but it is difficult to imagine circumstances in which someone would copy Cotrugli’s list of provinces onto a mappamundi unconnected with Cotrugli and that just happened to be in a style similar to his. The conclusion has to be that the mappamundi in Ottob. lat. 1417 came from a manuscript of Cotrugli’s De navigatione.

Insofar as it did contain a mappamundi, the manuscript of the De navigatione from which the map now in Ottob. lat. 1417 was taken would seem to have been closer to Cotrugli’s original manuscript than to either the Yale or the Schoenberg manuscript. The map suggests some inferences that can be made about that original manuscript. To start with, the folio on which the map is painted is larger (28.2 x 21.7 cm, or 11 x 8.5 inches) than the folios of either of the other two manuscripts. The difference in size could suggest that the small dimensions of the Yale and Schoenberg manuscripts represent subsequent evolution rather than the format of the author’s original manuscript. Furthermore, the arrangement of the lists of provinces around the mappamundi differs markedly from those in the Yale and Schoenberg manuscripts, where the provinces are listed before the map. It is tempting to see in the separation of the lists from the map an adaptation to the smaller format of these manuscripts. Thus the Ottob. lat. 1417 map not only indicates the geography of Cotrugli’s original map, but also gives clues about the nature of the original manuscript.

An exemplar of Cotrugli’s mappamundi that was added to a manuscript of Justin’s Historiarum Philippicarum libri XLIV, with Cotrugli’s lists of provinces in the margins. Oriented with South at the top, Vatican City, Biblioteca Apostolica Vaticana, MS Ottob. lat. 1417, fol. 2v.

The Disconnect between Map and Text

There is excellent evidence, then, that the map in Ottob. lat. 1417 comes from a manuscript of Cotrugli’s De navigatione. Yet a comparison of the map with Cotrugli’s text reveals significant disagreements, and these disagreements must be considered in order to achieve a full understanding of the relationship between map and text.

One of the most striking differences between the Ottob. lat. 1417 map and the description of the world Cotrugli borrowed from Ptolemy is that, whereas Cotrugli says that the Indian Ocean is enclosed by land, the mappamundi shows the Indian Ocean communicating with the circumfluent ocean.

Cotrugli’s De navigatione, Book 1, chapter 49: ‘. . . così dicono delo mare Indico, che tucto con li sui golfi Arabico, Persico, Gangetico, et de quello lo quale se chiama Golfo Magno, et de terra se chiude da ogni banna . . . ’ [they say the same of the Indian Ocean, which together with all its gulfs—the Arabian, Persian, Gangetic, and that which they call the Great Gulf—is encircled by land on every side].

In addition, no attempt was made to locate on the map all of the approximately sixty provinces carefully listed beside it (see Appendix). The shape of the British Isles is extravagant and does not resemble their outlines either on contemporary nautical charts or on the Ptolemaic maps that were beginning to circulate in the Latin West—although Cotrugli was evidently familiar with both types of maps. The mappamundi shows an island just outside the Strait of Gibraltar, probably intended to represent the Pillars of Hercules, but Cotrugli says nothing about such an island in Book 1, chapter 26, where he talks about the Strait. Compare the island on Ottob. MS 1417 just outside the Strait of Gibraltar with that on Andreas Walsperger’s mappamundi of 1448 (#245). In Book 1, Chapter 36, Cotrugli says that the Red Sea is disteso curvato ad modo de uno corno, or curved like a horn, but it does not show much of a curve on the mappamundi.

Other distinctive features of the map also lack any basis or corresponding descriptive passage in Cotrugli’s text. The Tanais River is depicted as flowing by a northern route between the (unlabeled) Sea of Azov and the Caspian Sea, an unusual but not unprecedented course, and in his text Cotrugli says only that the river flows out of the Sea of Azov. For example, fluvial connections between the Caspian and the Sea of Azov are found on Giovanni Leardo’s world map of 1452 or 1453 (#242) and on the Catalan Estense mappamundi of 1460 (#246). This conception of the Tanais River corresponds with that in 9th and 10th century Arabic geographical literature. In northern Asia, Cotrugli has a U-shaped mountain chain opening to the north, with a river flowing north from a city at the base of the ‘U’. This is one of only two cities labeled on the map, the other being Jerusalem. The mountains are probably supposed to be the Hyperborean. Cotrugli does briefly mention these mountains, but he says nothing about either the city or the river. The peninsula jutting east from the southeastern tip of Africa is probably a vestige of the Ptolemaic land bridge that joined southeastern Africa with southern Asia, enclosing the Indian Ocean, but Cotrugli does not allude to this feature.

Another characteristic of the Ottob. lat. 1417 map that is surprising in light of its apparent derivation from a manuscript of Cotrugli is its orientation. No labels on the map indicate the cardinal directions, but most medieval mappaemundi are oriented either to the East (the vast majority) or to the South, and particularly in a work about navigation that has two wind roses, we would expect the map to be oriented to one of the cardinal directions, and particularly to one of these two. The cartographer seems to have intended to place South at the top, but in fact it is oriented to the South-Southeast: if it were oriented to the South, we would expect the Strait of Gibraltar to be at the right-hand edge of the map, but the Strait is below the map’s horizontal midline, as if the map has been rotated 10 or 15 degrees clockwise. The apparent mis-alignment is brought out when the map is compared with the map of the world by Pietro Vesconte in Marino Sanudo’s Liber secretorum fidelium crucis (British Library, MS Add. 27376, fols 187v–188r) made c.1330 (#228). In this map, which is oriented to the East, the Mediterranean, which is barely visible in the inner margin, runs right down the center, and the north–south axis of the map appears nearer the center of Asia and Africa than it is on Cotrugli’s map.

What are we to make of the poor agreement between Cotrugli’s text and the mappamundi in Ottob. lat. 1417 that seems to have come from a manuscript of his work? Very little, Van Duzer would suggest. Given that only one folio was left for the map in the Yale and Schoenberg manuscripts, it certainly seems that Cotrugli was intending that his text would be illustrated with a mappamundi, rather than a Ptolemaic map, as is strongly confirmed by the existence of the Ottob. lat. 1417 mappamundi with Cotrugli’s list of provinces in the margins. In view of 15th century cartographers’ willingness to mix cartographical types, it would seem that from Cotrugli’s point of view, there was no real inconsistency in his use of a mappamundi to illustrate his Ptolemaic description of the world—even a mappamundi to which no changes were made to fit its context. Although two 15th century mappaemundi do show the Indian Ocean enclosed by land (for example #255), as Ptolemy indicated, they are not particularly Ptolemaic in any other respect, and it is not surprising that Cotrugli did not come across or devise something like these apparently rare transitional types. It is worth recalling that Fra Mauro (#249) discusses the impracticality of using Ptolemy’s first century system for depicting the 15th century world in his circular mappamundi of 1459.

It is quite clear that Cotrugli borrowed or copied his mappamundi from another source, much as he borrowed his description of the world from Ptolemy, rather than designing it himself. This was a common practice and should not surprise us any more than the inconsistency between the map and the text it was intended to accompany. For instance, the early 13th century mappamundi in Florence, Biblioteca Nazionale Centrale, Cod. Conv. Soppr. J. V. 6. (San Marco 189), folio 1v, precedes an extract from the Theorica planetarum generally attributed to Gerard of Cremona (fols 2r–2v). The map and the extract are part of the same gathering in the manuscript, but the map is of a type that illustrates some manuscripts of the pseudo-Aristotelian treatise De causis proprietatum elementorum. A 13th century manuscript of Walter of Chatillon’s poem Alexandreis (Paris, Bibliotheque nationale de France, MS lat. 8352), has a mappamundi on folio 100v that differs from most of the other maps in manuscripts of his work. The surface of the ecumene is covered with twelve large images of buildings that represent cities, but few other geographical details—a style much more typical of maps that illustrate manuscripts of Sallust’s De bello jugurthino. Other examples of this importation of mappaemundi from one manuscript tradition to another are not difficult to find, and in fact we will examine two additional instances of this phenomenon just below. But the occurrences have been cited should suffice to show that Cotrugli’s borrowing a mappamundi that did not fit well in the context of his work is not unusual.

Two Mappaemundi Copied from Cotrugli

Now that it has been established that the map in Ottob. lat. 1417 comes from a manuscript of Cotrugli’s work, we can identify two other medieval maps as having been copied from maps in the De navigatione, thus providing important new evidence for the diffusion and influence of that work. The first of the two mappaemundi is one whose source has long puzzled scholars. Only one of the extant sixty-four or so manuscripts and fragments of Cristoforo Buondelmonti’s Liber insularum archipelagi (c.1420), a book about the Aegean Islands that is illustrated with maps, contains a map of the world. This is MS Hamilton 108 in the Berlin Staatsbibliothek, which is thought to have been made about 1460– 1470. The map is on folio 81r. It has been reproduced a few times and briefly discussed, but no great light has been shed on it. Since there was no tradition of illustrating Buondelmonti’s work with a world map, this mappamundi must have been either created especially for this manuscript or copied from another source.

The only mappamundi in a manuscript of Cristoforo Buondelmonti’s Liber insularum archipelagi that was copied from Cotrugli’s mappamundi.

Berlin, Staatsbibliothek zu Berlin-Preussischer Kulturbesitz, MS Hamilton 108, fol. 81r, c. 1470

Comparison of the map in Hamilton 108 with that in Ottob. lat. 1417 reveals close similarities in terms of overall design, geographical details, toponyms (see Appendix), and the unusual South-Southeast orientation. Van Duzer argues then that the Buondelmonti map was copied from one in a manuscript of the De navigatione, although probably not the map that is now in Ottob. lat. 1417. The derivative nature of the Buondelmonti map is indicated by the comparatively careless handling of certain details: the fact that Ireland is missing, the faulty writing of the toponym cairo, and the vague courses of the Nile and Tanais, for example. Cardinal directions seem to have been added (there are none on the Ottob. lat. 1417 map), which have the effect of emphasizing the map’s curious orientation.

As the Appendix shows, the map in MS Hamilton 108 contains fewer toponyms than does MS Ottob. lat. 1417. This difference, together with the latter map’s more careful handling of some geographical details, suggests that the Vatican map is closer to Cotrugli’s original. Yet Ottob. lat. 1417 is not without errors. The traditional triangular shape of Sicily, shown correctly on the Hamilton 108 map, has been mistakenly applied to Crete. One is reminded of Ptolemy’s warning in his Geography 1.18.2-3 about the errors introduced during the copying of maps: After all, continually transferring [a map] from earlier exemplars to subsequent ones tends to bring about grave distortions in the transcriptions through gradual changes. As already noted, the shape of the British Isles on the Ottob. lat. 1417 map is highly unusual; on the Hamilton 108 map it is, if somewhat indistinct, at least not so strikingly strange. Such details indicate that the MS Hamilton 108 map is not as close to the map in Cotrugli’s original manuscript as the Ottob. lat. 1417 map. MS Hamilton 108 has been dated to 1460–1470, but given that Cotrugli composed his De navigatione in 1464–1465, the earliest possible date of MS Hamilton 108 should be brought forward at least into the second half of the decade.

The choice of Cotrugli’s mappamundi as a supplementary illustration for Buondelmonti’s Liber is a reasonable one. The addition of a world map lends the work a broader scope and thus a feeling of greater authority. Perhaps Cotrugli’s De navigatione was simply at hand when the copyist was creating the new manuscript, but Cotrugli’s mappamundi is detailed enough to show at least a couple of the islands that appear in Buondelmonti’s work, namely Rhodes and Crete, something that is not true of the majority of medieval mappaemundi, so the map effectively places the subject of Buondelmonti’s book in a global context.

The second mappaemundi bearing a remarkable similarity to that in Ottob. lat. 1417 is little known and all but unstudied. It appears in a work titled Ad Illustrissimum religiossimumque Portugalliae regem Emanuelem Iohannis Francisci Poggii Florentini Panegiricus [The panegyric of the most illustrious and devout King of Portugal, Manuel, by Gian Francesco Poggio [Bracciolini] of Florence], now Florence, Biblioteca Medicea Laurenziana, MS Ashburnham 1077. The work is by Giovanni Francesco Poggio Bracciolini (1447–1522), son of the famous Italian scholar and traveler of the same name, who lived from 1380 to 1459. Folio 24r of the manuscript contains a reference to the Portuguese conquest of Toprabana [Taprobana], which did not take place until 1505 or 1506. This date is a useful a terminus post quem for the making of the manuscript. The world map is on what is sometimes referred to as folio 1v, but elsewhere as folio 4v. Its execution has been attributed to Giovanni Boccardi (Boccardino Vecchio) on the basis of an inscription on the flyleaf that indicates that it was made by the same artist who painted the Breviary of Leo X. The only discussion of the mappamundi in Ashburnham 1077 is that of Destombes, Mappemondes, 180, no. 51.4, who transcribes the inscription on the flyleaf but does not mention Giovanni Boccardi.

The Ashburnam 1077 map is similar to both Cotrugli’s in Ottob. lat. 1417 and that in Hamilton 108. Notable among the parallels are the overall style, the configuration of most of the mountains and rivers, including the U-shaped mountain range in northern Asia with its northward-flowing river, and the Nile rising in a mountain range that is close to the southern coast of Africa and that extends eastward on a peninsula projecting into the Indian Ocean. Yet not everything is the same. The Ashburnham 1077 map has North at the top and entirely lacks both toponyms and city signs. On the other hand, it does have small black ships in the southern ocean, which are not present on the other two maps. They are found from the Strait of Gibraltar, along the western and southern coasts of Africa, near (unlabeled) Taprobana, and in the eastern part of the Indian Ocean. The allusion to the Portuguese explorations and conquests that are the subject of the Panegyricus contained in the manuscript is clear.

A mappamundi in a manuscript praising Manuel I, King of Portugal for his country’s explorations and conquests in Africa and asia. The map was copied from Cotrugli’s, but without the toponyms and re-oriented by the artist to have North at the top.

Florence, Biblioteca Medicea Laurenziana, MS Ashburnham 1077, fol. 4v

In contrast to Ottob. lat. 1417, some geographical details have been carelessly represented on the Ashburnham 1077 map. For example, on the maps in Ottob. lat. 1417 and Hamilton 108, the Tanais River takes a northern route between the Sea of Azov and the Caspian Sea through a break in the Riphaean mountain range, but on the Ashburnham 1077 map those mountains interrupt the river’s course. There are omissions too. Although the Ashburnham 1077 map is damaged in the area showing the north Atlantic, it is clear that Ireland was never included, a telling detail. Since Ireland is missing also from the Hamilton 108 map, it is tempting to hypothesize the existence of a branch of somewhat simplified maps that lack Ireland in the stemma of Cotrugli’s map.

A Question about Sources

One question remains to be addressed. The Hamilton 108 and Ashburnham 1077 maps seem to be evidence of the diffusion and influence of Cotrugli’s De navigatione, but another possibility must be considered, namely that these other two maps were copied not from Cotrugli’s map, but directly from the unknown source from which Cotrugli took his map. Although it is impossible to be certain about this, it can be shown that it is more likely that the two maps were copied from maps in manuscripts in Cotrugli’s work, rather than from the map that Cotrugli used as a source.

First, we should consider the nature of the two books in which the maps similar to Cotrugli’s appear and ask how likely is it that a manuscript of Cotrugli’s De navigatione would have been available to the copyists creating them. One of the books, Buondelmonti’s Liber insularum archipelagi, contains maps of the Mediterranean islands and describes a voyage among them, and the copyist executing it might well have had access to Cotrugli’s De navigatione. The second is an encomium of Manuel I, king of Portugal (1469–1521), in which the Portuguese explorations and conquests in Africa and Asia are praised. Again, since the subject of the encomium relates to navigation and the author is Italian, it is reasonable to suppose that the copyist would have had access to a manuscript of Cotrugli’s work. Unfortunately we do not know what work Cotrugli’s model map appeared in, and it is possible that these two artists would also have had access to that work. But the fact the artists’ use of Cotrugli’s De navigatione is consistent with the manuscript contexts of the two other maps at least renders plausible their use of that source.

But there is another, stronger argument that the two maps are copies of Cotrugli’s rather than of the map that Cotrugli used as a model. No manuscript of the work that contained the map Cotrugli used as a model has survived, whereas two manuscripts of Cotrugli’s text and three exemplars of his map are extant. It stands to reason that Cotrugli’s opus was more widely diffused than the book with the model map, and the better distributed work is more likely to have been the source of the Hamilton 108 and Ashburnham 1077 maps.

Benedetto Cotrugli’s De navigatione holds a special place in the history of instruction in navigation as the earliest surviving European treatise of its type. It would be tempting to assume that as such, Cotrugli’s book contributed to the diffusion of knowledge of this important science, but until now we have had only two manuscripts of the work, and little evidence of its dissemination or influence. This study has brought to light three examples of Cotrugli’s missing mappamundi, thereby in effect more than doubling the evidence for the work’s diffusion and demonstrating that in the 15th and 16th centuries it was considered a reliable source. In the process, we have been able to identify the source and show the relationship of three heretofore puzzling mappaemundi. Finally, light has been shed on the artists’ practice of copying a mappamundi from one work into another—of cartographic cross-pollination, so to speak—during the 15th and early 16th centuries.


University of Pennsylvania, Lawrence J. Schoenberg Collection, MS LJS 473, fol. 8r

Vatican City, Biblioteca Apostolica Vaticana, MS Ottob. lat. 1417, fol. 2v.

Berlin, Staatsbibliothek zu Berlin-Preussischer Kulturbesitz, MS Hamilton 108, fol. 81r, c. 1470


Destombes, Marcel, Mappemondes, A.D. 1200–1500 (Amsterdam, N. Israel, 1964), 187–88, no. 51.34 with plate 22.

Edson, Evelyn, Mapping Time and Space: How Medieval Mapmakers Viewed Their World.

Falchetta, Piero, ‘Il trattato “De navigatione” di Benedetto Cotrugli (1464–65). Edizione commentata del ms. Schoenberg 473. Con il testo del ms. 557 di Yale’, Studi Veneziani 57 (2009): 15–335, esp. 19–25.

Van Duzer, Chet and Sandra Saenz-Lopez Perez, ‘Tres filii Noe diviser- unt orbem post diluvium: the world map in British Library Add. MS 37049’, Word & Image 26:1 (2010): 21–39.

Van Duzer, Chet, “Benedetto Cotrugli’s Lost Mappamundi Found Three Times”, Imago Mundi, Volume 65, Part 1:1-14 (2013).