Waldseemüller’s Ptolemaic map of India.
Modern map of the Pakistani coast.
Though Reinal’s map of India is mostly superb for its time, the north-western part is very inaccurate between the Persian Gulf and the Indus river because it closely follows the old Ptolemaic model – rather than the mysterious and distinctly non-Ptolemaic source which I speculate was responsible for the rest of India’s coastlines on the Cantino and Reinal maps. That is why Reinal repeats the false north-western peninsula that is shown on Ptolemaic maps (such as Waldseemüller’s 1507 shown above).
The position, shape and orientation of the false Ptolemaic ‘Camba’ peninsula, as shown on the Reinal, plus the little island beside it, correlate well with the peninsula on which the modern city of Karachi is situated, although the scale is vastly exaggerated. This exaggeration may have originated in the reports of Alexander the Great’s sea-captain, Nearchus, who sailed back from the Indus towards the Persian Gulf and made specific mention of coastal features and a supposedly ‘haunted’ island along the way.
Reinal’s map of India, AD 1510.
The Reinal map departs from the Ptolemaic model specifically at the Indus delta (where Alexander stopped and turned back for home) and then southwards along the entire Indian coastline. As I said before, this coastline is infinitely more accurate than the Ptolemaic model, and strongly suggests that the source from which it is derived was far superior to anything previously available to Western seafarers and mapmakers. This coastline also correlates extremely well with Milne’s inundation maps showing India’s coastlines before about 12,000 years ago.
Of particular note is Reinal’s depiction of four small groups of islands, all close to India’s shoreline and all south of the NW bulge that should have been the Kathiawar peninsula. No such islands exist today, but Milne’s maps suggest that there were islands – including one very large one – in roughly the same positions, down to about 10,000 years ago.
India’s coastlines as they were in 11,500 BC.
Is it possible that what Reinal depicts are the remnants of these islands in the terminal stages of their post-glacial inundation?
Three of the island groups he shows lie along India’s west coast, in the right area for such remnants, and one lies immediately next to the southern ‘fish-lip’ (now less clear than at the LGM) at the very tip of the sub-continent.
Coasting the Indian Ocean
The Cantino and Reinal maps of the Indian Ocean were produced in an epoch of intense competition for trade and a real hunger for geographical knowledge on the part of the European powers that had witnessed – among many other breakthroughs – the rounding of South Africa’s Cape of Good Hope by Bartolomeu Dias in 1488,1 the ‘discovery’ of the Americas by Columbus in 1492, and the Portuguese encounter with the East that began when Vasco da Gama reached Calicut in south-west India in 1498.2
This first European crossing of the Indian Ocean was made from the East African port of Malindi (on the Swahili coast of modern Kenya) where da Gama and his small fleet arrived on 14 April 1498.3 There they were welcomed by the local chief, who arranged the services of ‘a loyal and extremely competent pilot’, Ahmed-bin-Majid, described as ‘the most famous expert in the navigation of the Indian Ocean in the 15th century’.4 With this man as their guide they reached India very rapidly, anchoring in front of Calicut on the Malabar coast on 20 May 1498.5
There the Portuguese remained for several months, attempting to put arrangements in place to build a trading post, but were foiled at every step by established Arab merchants alarmed at the prospect of European competition ruining their business with the East. Eventually da Gama left empty-handed, ‘convinced that only a stronger expedition … would have the power to bring negotiations to a successful conclusion’.6
On the return voyage there were outbreaks of scurvy, often as few as half a dozen crew were well enough each day to man the ships, and the fleet was alternately becalmed then driven off course by contrary winds. Their zigzag route took them through the Lacadive archipelago – which da Gama named the Santa Maria islands – and then to the small island of Angediva, some 70 kilometres south of Goa.7 Many died during the crossing to Malindi, which took three times as long as the outward passage, and it was the summer of 1499 before the survivors limped home to Portugal in their two remaining ships.8
Almost immediately after da Gama was welcomed back by King Manuel, the Portuguese monarch announced that a new, armed fleet would be sent to India – thirteen ships, with crew and soldiers totalling 1500 men, under the command of Pedro Alvares Cabral. Such a force, it was felt, would be sufficient to set aside the political and commercial obstacles that had confronted da Gama.9
The new fleet set sail on 9 March, reaching the Canary islands five days later and the Cape Verde islands on 22 March 1500. There one of the ships was ‘eaten by the sea’.10 The remaining twelve crossed the Atlantic to South America where Cabral made landfall in Brazil on 26 April, claiming it for Portugal. Sending one ship back to Lisbon with news of the discovery of the land that was first known as Vera Cruz, then later Santa Cruz, and finally Brazil,11 he remained only until 2 May, then turned his fleet south-east and set a course for the Cape of Good Hope.12
Voyages of Vasco da Gama and Pedro Alvares Cabral, AD 1498–1500.
By this point Cabral’s fleet was reduced to eleven ships. Rounding the Cape of Good Hope four more vessels went down with all hands in a violent tempest – among those who perished was Bartolomeu Dias, who thus ‘came to be buried in the waters of which he had been the discoverer twelve years before’.13 A fifth ship, separated from the fleet in the same storm, discovered the island of Madagascar, and then returned to Portugal on its own.14
Cabral was therefore down to six ships and less than half his original force when he crossed the Indian Ocean to Calicut. The opposition to a Portuguese trading post still remained strong there and he was now no longer in a position to overcome it. He therefore sailed further south along the Malabar coast looking for a friendlier reception and found it at Cochin where the local rajah permitted him to set up a ‘factory’. Cabral then took the fleet to Cananor where they loaded cargoes of spices before returning to Portugal in the early summer of 1501, just over a year after they had left.15
Although in both cases under extreme time pressure and in difficult circumstances, the expeditions of da Gama and Cabral undoubtedly did conduct some cursory exploration of several hundred kilometres of the Malabar coast between roughly 15 degrees north latitude (Goa) and roughly 10 degrees north latitude (Cochin). On the third and fourth expeditions, however, these explorations were not extended:16 ‘It was only with the fifth India fleet in 1503 under Albuquerque that exploration was carried further, as far as Coulon [Quilon], almost on the southern tip of Malabar.’17
Cape Comorin – modern Kaniya Kumari, the true southern tip of the Indian peninsula – was first rounded near the end of 1505 by a fleet under Lourenco de Almeida. The fleet had been sent to the Maldives to spy on the sea trade with the Indonesian islands further east but was carried off course to Cape Comorin by winds and currents. From there Almeida sailed his ships to Sri Lanka: ‘Thus Lourenco de Almeida and his companions were the first Portuguese to pass into the eastern Indian Ocean.’18
In 1506 there was another ‘first’ – Joao Coelho was the first Portuguese to reach the northern terminus of the Bay of Bengal and ‘to drink the waters of the Ganges’.19 But it was not until 1509 that Diogo Lopes de Sequeira made the first full crossing of the Bay of Bengal to reach Malacca20 – the Malaysian peninsula known until that time on Ptolemaic maps as Aurea Chersonesus, the Golden Chersonese.21
Thus, it can be seen that the focus of Portugal’s attention for more than a decade after Vasco da Gama first reached India in 1498 was on the Malabar coast south of Goa and on the eastern Indian Ocean and the Bay of Bengal. The long lines of supply and relative scarcity of men and ships meant that no attention could be paid to the stretch of the Indian coast tha
t runs north-westwards from Goa, at roughly 15 degrees north latitude, past the Gulf of Cambay, the prominent Kathiawar peninsula and the mouths of the Indus, up to the northern terminus of the Arabian Sea at roughly 25 degrees north latitude. As Damiao Peres writes in his authoritative History of the Portuguese Discoveries:
In the first years of Portuguese expansion in the Indian Ocean, the reconnaissance of the Gulf of Arabia [i.e., the Arabian Sea] was limited to a few southern ports of the Malabar coast to the east, and to the coast of Arabia and its neighbouring areas to the west. Included were some island groups lying between the two. The northern part of the Gulf of Arabia [Arabian Sea] and its adjacent waters – the Persian Gulf and the Red Sea – were only visited in the first years of the second decade of the sixteenth century.22
The mystery of the Cantino map of 1502
And this brings us to what is mysterious about the Cantino map – so named after Alberto Cantino, the Lisbon-based diplomatic agent of the powerful Duke of Ferrara in Italy.23 Cantino somehow acquired this beautiful but unsigned world map in Portugal, or had a cartographer there copy it specially from another map, and then smuggled it out of the country, getting it to Italy by or before 19 November 150224 (no mean feat, since Portugal was jealous of its discoveries and imposed the death penalty on those caught smuggling maps out of the country).25
India on the Cantino planisphere of c. AD 1502.
When was the Cantino map drawn?
Let us start by stating the obvious: it must have been drawn before 19 November 1502, when it reached Italy. Indeed, according to H. Harisse, it typically took craftsmen of the period about ten months to prepare such a map. If this is correct, then it pushes the origins of the Cantino back at least to the beginning of 1502.26
Going back still further, there is internal evidence in the map itself which proves that it could not have been drawn much before the summer of 1501. That was when the ships of Cabral’s second India fleet returned to Portugal from their voyage – begun a year previously – that had taken them not only to India but also to South America. The evidence survives because the Cantino is a world map that shows – and claims with Portuguese flag-icons – the section of the Brazilian coast discovered by Cabral in 1500.27 Since similar flag-icons are likewise seen over Cochin and Cananor in southern India – which Cabral reached later in 1500 – it is the inescapable conclusion that the Cantino map expresses knowledge that could only have been acquired on the Cabral voyage.
Indeed this is the conclusion of orthodox historians of cartography,28 so it is not controversial to restate it here. What is extremely strange, however, is that neither on Cabral’s voyage of 1500/01, nor on the earlier 1498/9 voyage of Vasco da Gama, nor on any later Portuguese voyage until after 1510, was the north-western part of India ever visited. Yet the Cantino map shows north-western India very clearly. And although the portrayal is inaccurate vis-à-vis India’s western coast as it has looked for the past 7000 years – in the single, significant respect that it entirely omits the Kathiawar peninsula – it is still hugely more accurate for India as a whole than the grotesque image of the subcontinent provided in the Ptolemaic maps.
Particularly noteworthy is Cantino’s representation of the east coast of India. In general (see diagrams) it matches well to what the east coast of India should look like.
I do not deny that the Portuguese were capable of drawing maps as accurate – and indeed more accurate – than this one. But the puzzle to me is how Cantino’s Portuguese cartographer could have acquired such accurate knowledge of the outline of eastern India as early as 1501–2, when historical records show that the fleet of Lourenco de Almeida did not even round Cape Comorin and enter the eastern Indian Ocean until 1505? This part of the map also shows Sri Lanka at close to its correct size and very close to its correct location more than three years before Lourenco de Almeida became the first Portuguese to sight Sri Lanka.
Surely, therefore, curiosity should drive us to find an explanation for the existence of this strikingly good chart of supposedly uncharted waters?
‘T-O’ maps
Good is a relative term. To understand why the Cantino and Reinal maps of the Indian Ocean are ‘good’, and in fact in some ways close to ‘revolutionary’, we need to view them in the cartographic context of their place and time – i.e. Europe and the Mediterranean in the fourteenth to the sixteenth centuries AD.
The Augsburg T-O map of AD 1472.
During this period mariners, merchants, adventurers and armchair travellers had at their disposal four distinctly different types of maps and charts. The simplest of all – far too simple to be of any use to navigators – are the so-called ‘T-O’ maps. With a long history going back to the seventh century AD, these show an encircling ‘O’ of water that is often inscribed with the words ‘MARE OCEANUM’ -representing the ‘Ocean Sea’ (sometimes ‘Ocean River’) that was believed in antiquity to surround all the lands of the world29 (an idea, by the way, that is completely correct, as all the world’s oceans do indeed interconnect). Inside the ‘O’ a ‘T’ is then inscribed, dividing up the land into the three known continents of Africa, Asia and Europe. The vertical stroke of the T represents the Mediterranean, separating Africa from Europe and adjoining the Ocean Sea at the Atlantic. The cross-bar of the T is the north-flowing river Nile on one side of the Mediterranean, the south-flowing river Don on the other side of the Mediterranean and also, vaguely, the Black Sea, the Bosporus and eastern Mediterranean, beyond which lies the continent of Asia. The Garden of Eden is also often depicted on the ‘top’ of such maps, which are oriented eastwards. Map historian John Goss points out that frequently ‘Four rivers were also described as flowing from the Garden of Eden: Psihon, Gihon, Tigris and Euphrates.’30
The T-O maps provide at best a ‘shorthand picture of the world’.31 But the enduring power and pervasiveness of this essentially useless cartographic tradition is illustrated by the oldest surviving printed map of Europe – a T-O map, printed by Gunther Zainer at Augsburg in 1472, that reproduces exactly the original concept as set out by Isidore, Bishop of Seville, in his Etymologiarum written in the early seventh century.32
Mappamundi
The second category of maps and charts available between the fourteenth and sixteenth centuries is known as the mappamundi. It is important to be clear that this is a very distinct and specific type of world map (because, in the texts of those times, other world maps of completely different types were also sometimes referred to as mappaemundi or mappamundi, when what was meant was just ‘world maps’ in the rather loose, general sense of ‘maps of the world’).33 So, to be clear, the mappamundi to which I refer here were normally hand-painted on cloth or vellum (hence the origin of the name mappamundi — meaning, literally, ‘Cloth of the World’). The classic example is the Hereford mappamundi, attributed to Richard of Haldingham C.AD 1290, but mappamundi continued to be made well into the fifteenth century. They retain the essential design of the T-O maps but greatly increase the amount of detail concerning mountains, rivers, pilgrim routes, etc., on the three recognized continents of Africa, Asia and Europe – sometimes taking into accounts myths, legends and recent traveller’s tales. Unfortunately, none of the specifically geographical details that these maps provide would have been of the slightest bit of use to travellers or mariners since all the details – all of them! – are wrong, misguided and misleading.34 In short, the mappamundi promulgate a wholly incorrect image of the world – an image that is almost all dry land and that reduces the Ocean Sea covering seven-tenths of our planet to the narrow, ribbon-like rim of the surrounding ‘O’. ‘The very crudeness of the geography of the Hereford map’, comments John Goss, ‘reflects a marked deterioration in geographical knowledge from the time of Ptolemy a thousand years earlier.’35
Hereford mappamundi, c. AD 1200.
Ptolemaic maps
Almost nothing is known about the life of Claudius Ptolemy.36 His first name is Roman and his second Macedonian.37 He is thought to hav
e been born in Upper Egypt38 C.AD 90 and to have died around 168.39 A scholar at the Library of Alexandria from roughly AD 127 to 145,40 his two famous surviving works are the Almagest (Ho megas astronomos), a book of astronomy and cosmology in which he expounds the ‘Ptolemaic system’ of a fixed spherical earth at the centre of a revolving universe, and the Geography (Geographike hyphegesis), in which he includes information on how to construct maps of places in Europe, Africa and Asia tabulated according to latitude and longitude.
It is not absolutely clear whether Ptolemy ever in fact drew maps himself, or even had maps drawn to accompany his work.41 Strictly speaking, they were not necessary because his primary method was to provide the longitude and latitude coordinates of more than 8000 places and topographical features in such a way that: ‘the reader can draw for himself regional maps on various suitable scales, and even a general map of the world’.42