This html article is produced from an uncorrected text file through optical character recognition. Prior to 1940 articles all text has been corrected, but from 1940 to the present most still remain uncorrected. Artifacts of the scans are misspellings, out-of-context footnotes and sidebars, and other inconsistencies. Adjacent to each text file is a PDF of the article, which accurately and fully conveys the content as it appeared in the issue. The uncorrected text files have been included to enhance the searchability of our content, on our site and in search engines, for our membership, the research community and media organizations. We are working now to provide clean text files for the entire collection.
By Commander Bruce A. Castleman,
U.S. Navy
There were some great 15th century navigators: the Portuguese, backed by a large hydrographic establishment, and the Arabs, heirs to centuries of tradition. But all things considered, Christopher Columbus- born in Genoa and sailing for Spain—was first among equals.
major advances in the navigator’s art between 1500 and their famous voyages.
Navigation in Columbus’s time consisted of two basic techniques: dead reckoning and celestial navigation. Dead reckoning (DR) is based on starting from a known point and then laying down a series of vectors displaying courses and distances travelled. Modern navigators plot DR legs on Mercator-projection charts, which have parallel lines of both latitude and longitude. Charts could not be used for plotting positions until the Mercator projection was developed in 1569, but the ones then available did show the coastlines, and major ports and landmarks were con- uld suffer unfairly in such a comparison because he far more behind for posterity’s criticism that did any er seaman.6 I agree with Morison, and so dead recx- ng and coastal piloting also figure in my comparison^ \ celestial fix is possible only if the navigator has an :urate timepiece, and these were not available unti 18th century.7 Earlier navigators were therefore lim- J to those special-case observations that do not re- re an accurate timepiece, and, even then-anr^lman^ itaining certain celestial data was required. There are ee special cases that will yield a latitude line, yn observation of the altitude of Polaris, the North star.
Astrolabes like this one measured angles of elevation, but depended on free-hanging plumb weights to maintain their vertical orientation. Observations from the decks of ships at sea often had errors of up to 5°.
with the latitudes of various destinations.17 Aligning the bottom of the marked leg with the horizon, the movable leg would then be pointed at the sun or star and would intersect the marked leg on the latitude scale.
The other instrument was called a kamal, it consisted of a rope, along which slid a rectangular piece of wood. The navigator held the rope in his teeth, stretched it out, and then moved the board until the bottom was on the horizon and the top just touched the heavenly body. The position of the board on the rope thus correlated to a line of position.18 Since the Arabs did not have chronometers, the kamal and the triangular quadrant must have been used only to take meridian altitudes to get latitudes. The latitudes of important places were listed in publications of sailing directions called rahmanis, which were in use before the 11th century.19 When Vasco da Gama arrived in eastern Africa en route to India, he found the kamal and the triangular quadrant to be in general use by the Arabs. The European cross-staff is functionally identical to the kamal, but it was not commonly used until about 1550.■" Ahmad ibn-Majid (c. 1435-c. 1500) is often held to be the greatest of the Arab navigators. The Portuguese called him Malemo Cana, corrupted from the Arabic for “sailing master,” and “master of astronavigation and of mathematics.” Ibn-Majid wrote many of the rahmanis and 19 of his manuscripts are held by the Bibliotheque Na-
if small corrections are applied to the observed altitude
- An observation of the sun’s altitude at noon when it crosses the observer’s line of longitude will yield a latitude if the sun’s declination is known. Solar declination varies over a range of 47° in 182 days, so the daily change is less than 1°. Since 1° was less than the instrument error during the period under discussion, an almanac with a table of solar declinations with only one value per day met the needs of Dias and Gama.
- An observation of the highest altitude of a star, as it crosses the observer’s meridian also yields latitude since stellar declinations are constant. Arab navigators may have used this method in the Indian Ocean, and the Portuguese
were developing a similar method during the 1490s.
Although rudimentary by today’s standards, celestial navigation in 1500 required fairly involved applications of mathematics and science, and only educated and literate sailors were capable of this.8 The Portuguese were by far the most advanced European practitioners of celestial navigation; as early as 1484,
King John II had appointed a Junta de Mathematicos (a panel of mathematicians) that developed a practical form of astronomy for use by marine navigators.9
Rabbi Abraham Zacuto of the University of Salamanca in Spain developed solar declination tables for the period 1473-1478, and his work was translated by his student, Rabbi Jose Vizinho, who was the Junta's astronomer. Viz- inho then developed a practical set of solar tables and checked them during a voyage to Africa in 1485.10
The Junta's publications were not originally intended for general use. The most important publication, the Regimento do Astrolabio e do Quadrante (Rules for the use of the Astrolabe and Quadrant), was issued—in manuscript form—only to Portuguese voyagers until Columbus’s discovery. It was first printed on a press in 1495," after Columbus had broken Portugal's virtual monopoly on exploratory voyages.
Fifteenth century navigators had difficulty in squaring celestial results with dead reckoning, because they did not know the length of 1° of latitude. They were superb at estimating boat speed and distance travelled, but without knowing that 1° equalled 60 nautical miles, they were confused when they compared DR with celestial altitudes. Columbus believed that 1° equalled 45 nautical miles, based on his incorrect assumption that the Arab astronomer Alfragan had used Roman miles to calculate his degree of 56.66 Arabic miles.12 Alfragan’s degree actually equates to 66 nautical miles. If the Junta believed Alfragan, or accepted Eratosthenes’s value of about 60 nautical miles, then the contention that King John II rejected Columbus’s proposal on technical grounds is more credible than the popular belief that Columbus’s demands for rewards were excessive.13
Navigation was thus a matter of dead reckoning, which the most advanced navigators attempted to double-check with latitude observations.14 Things have not changed much in that regard, for today, celestial observations are still basically used as a double-check.
The Arab navigator’s art must be examined separately. Arab astronomy was well-developed in its own right, and this was reflected in their navigation of the Indian Ocean. Beginning about 800, Arab navigators sailed the Indian Ocean from the Persian Gulf to Madagascar and beyond to China, using charts plotted on X-Y axes.15 Their charts were accurate, as can be seen by looking at the Indian Ocean coast on a chart in the 1375 Catalan Atlas of Mal- lorcan cartographer Abraham Cresques.16
This is evidence that parts of the Arab seafarers’ body of knowledge were known to Iberians prior to the 15th century voyages.
The Arabs had used astrolabes for several centuries, and had attempted to overcome the problem of the pitching deck by developing two instruments that used the visible horizon as a reference, rather than a plumb bob. One was a triangular quadrant, one of Whose legs would be marked
tionale in Paris.21
When Vasco da Gama arrived in the eastern African port of Malindi, the local sultan sent ibn-Majid along to pilot da Gama across across the Indian Ocean to India.22 Accounts of ibn-Majid reporting aboard da Gama’s flagship describe the Arab and the Portuguese comparing notes, charts, and instruments. Ibn-Majid was already familiar with Portuguese instruments and methods of celestial navigation, and he demonstrated the use of the kamal and the triangular quadrant. The only surviving first-hand account of Gama’s voyage is the Roteiro of Alvaro Veltio, who was a soldier and not a sailor, but who noted on 10 March 1498 that Arabs used compasses, quadrants, and navigating charts.23
Ibn-Majid’s service were of “inestimable value” to Gama;24 without the Arab’s help, he would have had a difficult time threading his way through the Laccadive Islands off the southwest coast of India.
It is easy to imagine an unforewarned Portuguese squadron losing every ship on the Laccadive reefs and never reaching Calicut.
Ahmad ibn-Majid was the equal of his most advanced European contemporaries in virtually every respect. With a thorough knowledge of win current, gained through many year! at sea before Gama’s arrival, he. must have been good at dead reck oning also. Throughout his life however, he sailed within a knows geographic area, since his fore runners for centuries had ranged as far as China. In this significant respect, he was outstripped by the other three mariners, each of whom voyaged far into unknown waters.
From Bartolomeu Dias no journal or logbook survives, so the quality of the charts drawn from his data are the only indicators of his ability as a celestial navigator. The charts of the western coast of Africa from the 1480s appear reasonably correct, and the stone crosses that Dias left behind, when compared with the positions he charted using astrolabe and quadrant observations, indicate that he was accurate to within about two degrees.25 These observations were taken ashore, so one cannot compare them with observations taken at sea that had errors of 4°. Only once did Columbus ever get that close.26
Dias’s determination of the latitude of the Cape ot Good Hope appears to have been very close, and when Gama made his landfall nine years later, he went ashore to check his latitude and obtained satisfactory results.' Columbus Was present in Lisbon when Dias returned, and recorded in a book margin that Dias reported the Cape ot Good Hope to be at 45° degrees south latitude—which is high by about 13°.28 Morison ascribed the error to Dias, but it is difficult to reconcile such an error with the excellent landfall made there by Gama after three months beyond the sight of land. Either Columbus erred, or perhaps the public disclosure amounted to disinformation by the Portuguese crown in order to maintain a veil ot secrecy over
their findings, as they did with nautical almanac data at the time.
Dias made many voyages, both in command and as ship’s pilot, although it must be remembered that his thorough exploration of the western coast of Africa was done within sight of land.24 He returned to Portugal from the Cape of Good Hope also using coastal piloting,5" but this method of navigation requires good seamanship; most groundings occur within sight of land, and Dias explored thousands of miles of uncharted coast without ever los ing a ship.
Dias’s actual rounding of the Cape of Good Hope is remarkable. After a long period fighting headwinds down to 29° south latitude, he sailed to the southwest, beyond the sight of land, until he found winds that allowed him to sail directly south into the west wind drift, and then eastward through the rough seas, turning north only when he was certain that he was east of the coast along which he had isailed.11 He landed on the Natal coast, and the gateway to India was open. This plunge into rough, uncharted seas paid off handsomely—from all indications, Bartolomeu Dias was a mariner with a great deal of sea sense. Christopher Columbus s
_ _____________ original log is lost, but Bar-
tolome de Las Casas produced an abstract of quotes and paraphrases that passes today for a log of the first voyage. Although numerous transcriptions and translations of the abstract have been made, many historians question their accuracy.32
Several observations of Polaris are recorded in Columbus’s journal, but there is no record that he ever took an observation of any other celestial body. Columbus first mentions his quadrant in his entry of 24 September 1492. He also had an astrolabe on board during the first voyage, but never used it, probably because he did not know how.31 After arriving in the Indies in 1492, Columbus too several quadrant observations, but they were gross > in correct—often by more than 20°—and he gave up on is quadrant, deciding it needed to be repaired when e go
back to Spain.34 „ ,
Morison developed an extensive argumentthat bus took his observations on the wrong star, but 1 find tms impossible to believe, especially since the error was repeated several times.35 More plausible is the later argument of James E. Kelley, Jr., who noted that Columbus s results corresponded to the arc-tangents of the observed altitudes, and that, since many quadrants of the day had arc-tangent scales, Columbus probably read the wrong scale on his instrument16 In any event. Columbus obviously discounted his observations because he knew he was
in lower latitudes; after all, he deliberately sailed due west from the Canary Islands in order to run before the northeast tradewinds.31
No records survive of Columbus’s celestial navigation on his second voyage. King Ferdinand and Queen Isabella suggested to Columbus that he take along an astronomer, but he did not comply. He did explicitly promise the sovereigns an accurate survey of his findings prior to sailing on his first voyage, and his failure understandably dismayed his employers.38
Columbus sailed on his third voyage to the Indies in May 1498. His celestial navigation skills had improved, but not to the point that he could achieve useful readings. He reported that many Polaris sights were taken while sailing westward along the 10th parallel of north latitude.™ He observed correctly that Polaris rotated around the actual celestial North Pole, and measured the radius of the circle as five degrees.40 The actual diameter was about seven degrees, so he was actually quite good for those days. Significantly, he made these accurate observations while at sea, not ashore or in a calm harbor. Unfortunately, he did not know the corrections to apply to obtain latitude. (Morison describes the procedure, called the “Rule of the Guards,” in detail in an article in The American Neptune.)[1]' Columbus did his best celestial work on his fourth voyage, when he was marooned in Jamaica; he determined that he was at 18° north latitude, which was correct.42
Columbus did not excel as a celestial navigator. He was self-taught, and did not have the technical support provided by the Portuguese crown to Dias and Gama. Unlike ibn-Majid, he was not heir to a centuries-old body of knowledge. Doubtless, he would have done much better with access to those resources, but to have done any celestial work at all singled him out from the average mariners of his day.
In dead reckoning and coastal piloting, Columbus shone. He knew the wind patterns along the coasts from Guinea to England,43 and he usually chose his transatlantic tracks to take advantage of the prevailing winds. On his first voyage, he started from the Canary Islands and ran west before the northeast tradewinds. On his return, he went north and found the westerlies that took him back to Spain. On the same return voyage, Columbus at one point estimated the fleet’s position to be west of the Azores, while the other pilots all estimated positions east of those islands. Three days later, the Nina made landfall in the Azores.
Columbus’s subsequent landfall in Lisbon was far more accurate than that made by Martin Alonso Pinzon in the Pinta, which had become separated from the fleet.44
Columbus was amazingly accurate on each of the rest of his ocean crossings. He planned his westbound crossing for the second voyage to make landfall at Martinique, whose location had been described to him by the Taino Indians when he was in Hispaniola on the first voyage. He actually made his landfall at Dominica, which is only a few miles north of Martinique. This outbound track on the second voyage became the standard for sailing ships for centuries afterward.45
His return voyage to Spain from this second crossing was equally successful as, after six weeks of zigzagging in light winds, Columbus made his predicted landfall at Cape St. Vincent. Why he selected a southerly course, rather than sailing north to seek the mid-latitude westerlies as he had done on the first voyage, is unknown; the logs that have come down to us give no reason. On the third voyage, he split his fleet, and sent half the ships on the route he had pioneered to Dominica; his instructions to these captains worked perfectly.46 The other half of the fleet proceeded farther south before turning to the west, then sailed across the Atlantic and made landfall at Trinidad. Columbus returned from the third voyage as a prisoner in chains, so there is nothing to report for that transit.
On the fourth voyage, storms pushed the ships slightly south of the intended track on the westbound crossing and the fleet’s landfall was Martinique, instead of Dominica. The storms gave the ships a boost, though, and the crossing took only 20 days.47
During each of his voyages, Columbus navigated around the reefs, cays, and coasts of the Caribbean with great skill. As J.H. Parry points out, Columbus was successful far too consistently to ascribe his landfalls and navigation to luck.48 He did lose a ship, though, when his flagship, the Santa Maria, ran aground on 24 December 1492. The Diario entries for the period show that this was the result of dereliction by his subordinates’ failure to carry out his standing orders, and that view was not refuted in any of the subsequent Spanish legal proceedings, nor has it been questioned by any modern historian.
On his fourth and final voyage, Columbus was marooned in Jamaica because teredo worms had so weakened his two ships that they were in danger of foundering, and he deliberately put both ships aground in Jamaica to avoid breaking up at sea.49 Of course, he erred in not careening the ships before they were so badly damaged.
Vasco da Gama (ca. 1460-1524) commanded the squadron that completed the Portuguese kings’ search for an ocean route to the Orient. He sailed from Portugal in 1497, nine years after Dias rounded the Cape of Good Hope. After fighting headwinds along the coast of Africa, he decided to sail south through the middle of the South Atlantic to the Cape, and reached St. Helena Bay on the African coast after 13 weeks at sea; at the time, this was the longest recorded period spent beyond sight of land by any European sailor.50 Morison, who had the highest regard for Columbus, ranked Vasco da Gama as a great navigator for daring to to go offshore to find favorable winds
and reduce the time to reach the Cape of Good Hope.51
He then bucked the winds for four days until an abnormal shift in the prevailing easterly winds allowed him to sail past the Cape.52 A lucky break rescued him from the possible consequences of his decision to attempt to round the Cape so close to land, which was ill-advised in view of Dias’s experience. The rest of the voyage as far as Malindi was well-executed, including reconnaissances of several river mouths.
As a celestial navigator, Vasco da Gama had the benefit of support from the Portuguese hydrographic establishment. He had specially made astrolabes, and used them in conjunction with the Regimento do Astrolabio e do Quad- rante53 That is far more than Columbus could do, but it seems likely that Dias would have had the same ability.
Ibn-Majid appears to have been the best celestial navigator, but he was the heir to a centuries-old tradition of knowledge, and apparently never ventured into unknown seas, whereas Dias took a long loop around the Cape of Good Hope and Vasco da Gama took a much longer sweep through the South Atlantic. Columbus was the weakest of the four in celestial navigation, but his coastal piloting performance was as good as either Portuguese mariner. Ibn-Majid’s survival to the ripe old age of 65 hints at a long record of prudent, successful seafaring.
Columbus’s remarkable record of dead-reckoning performance sets him apart from the others. He made more voyages than did Dias or Gama, and succeeded in each voyage without the benefit of either Arab tradition or the Portuguese hydrographic establishment. Although each of the four was a highly capable mariner in his own right, the evidence clearly puts Columbus above the rest. [2] *
M.J.L. Young, J.D. Latham, and R.B. Serjeant, eds. (Cambridge: Cambridge University Press, 1990), p. 325.
'’Velho, Alvaro, A Journal of the First Voyage of Vasco da Gama, translated by E.G. Ravenstein, commonly known as the Roteiro (London: Hakluyt Society, 1898 [1963 reprint]), p. 26.
“William Facey, Oman: A Seafaring Nation (Muscat: Ministry of Information and Culture, Sultanate of Oman, 1979), p. 99.
“George Fadio Hourani, Arab Seafaring in the Indian Ocean and Early Medieval Times (New York: Princeton University Press, 1951), p. 107.
“Morison, Admiral of the Ocean Sea, Vol. I, p. 258.
’'Henry H. Hart, Sea Road to the Indies: An Account of the Voyages and Exploits of the Portuguese Navigators, Together with the Life and Times of Dom Vasco da Gama, Capitao, Viceroy of India, and Count ofVidiguiera (London: Wm. Hope and Co., 1952), pp. 154-155.
"J.F.P. Hopkins presents a highly cogent argument that this was not really ibn- Majid in “Geographical and Navigational Literature," loc. cit., p. 326. The question is unimportant to this study, though; whoever he was, sources from the time unanimously praised the abilities of the Arab pilot who boarded Gama’s flagship.
If this was just an average colleague of ibn-Majid, then perhaps “the master’s" skill will have been underestimated, but my conclusions remain.
’’Velho, Roteiro, p. 26.
’’Parry, The Discovery of the Sea, p. 174.
’’Axelson, Congo to Cape, p. 45.
“Morison, Admiral of the Ocean Sea, Vol. 1, p. 242.
"Velho, Roteiro entry of 9 November 1497; and Axelson, Congo to Cape, p. 158.
"Morison, Admiral of the Ocean Sea, Vol. I, p. 99.
"Bell, Portugal and the Quest for the Indies, p. 124.
“Parry, The Discovery of the Sea, p. 132.
"Bell. op. cit., p. 151.
’’See Kirkpatrick Sale, The Conquest of Paradise: Christopher Colunbus and the Columbian Legacy (New York; Alfred A. Knopf, 1990), pp. 20-21, and David Henige’s articles "Samuel Eliot Morison as Translator and Interpreter of Columbus’ Diario de a Bordo," Terra Incognitae, Vol. 20, 1988, pp.69-88, and “Edited . . . and not Precipitated: Three Recent Editions of Columbus’ Diario de a Bordo,” Terra Incognitae. Vol. 22, 1990, pp. 93-109.
’’Morison, Admiral of the Ocean Sea, Vol. I, p. 241.
"Gianni Granzotto, Christopher Columbus: The Dream and the Obsession: A Biography, translated by Stephen Sartarelli (London: Grafton, 1988), p. 179. ’’Morison, “Columbus and Polaris, Part I,” The American Neptune, Vol. I, No. 1, January 1941, p. 22.
“James E. Kelley, Jr., “In the Wake of Columbus on a Portolan Chart,” In the Wake of Columbus: Islands and Controversy, Louis de Vorsey and John Parker, eds., (Detroit: Wayne State University Press, 1985), p. 108.
’’Morison, Christopher Columbus, Mariner, p. 38.
“Morison, “Columbus and Polaris, Part II,” The American Neptune, Vol. I, No. 2, April 1941, p. 123.
'‘Christopher Columbus, “Letter to the Sovereigns Concerning the Third Voyage." text in Spanish with English translation by Cecil Jane on facing pages. Select Documents Illustrating the Four Voyages of Columbus (London: Hakluyt Society, 1933), Vol. II, pp. 28-34. This letter also contains Columbus's preposterous “nipple and breast" theory about the shape of the world, consideration of which is beyond the scope of this paper.
‘“Columbus, “Letter to the Sovereigns Concerning the Third Voyage," Vol. II, p. 28.
“Morison, “Columbus and Polaris, Part 1,” PP- 13-25.
“Morison, “Columbus and Polaris, Part II,” pp. 134-135.
“Christopher Columbus, The Diario of Christopher Columbus's First Voyage to America 1492-1493, abstracted by Fray Barloiome de Las Casas, translated by Oliver Dunn and James E. Kelley, Jr. (Norman, Oklahoma: University of Oklahoma Press, 1989), pp. 356-357.
“John W. McElroy, “The Ocean Navigation of Columbus on his First Voyage'. The American Neptune, Vol. I, No. 3, July 1941. pp. 229,240.
“Morison, Admiral of the Ocean Sea, Vol. II, p. 64.
“Ibid, pp. 188-190, 236.
“Granzotto. op. cit., p.251.
“Parry, The Discovery of the Sea, p. 195.
“Morison, Admiral of the Ocean Sea, Vol. H, p. 383.
' Parry, The Age of Reconnaissance, p. 156.
“Morison, The European Discovery of America: The Southern Voyages 1492-1616 (New York: Oxford University Press, 1974), p. 222.
’’Velho, Roteiro entries for 18-22 November 1497.
’’Parry, The Age of Reconnaissance, p. 112.
Commander Castleman is the Commanding Officer. Naval Surface Force, Pacific Mobile Training and Evaluation Command, in San Diego, California. He has served as main propulsion assistant and navigator on the USS Robison (DDG-12), reactor mechanical assistant on the USS Enterprise (CVN-65), engineer officer on the USS Somers (DDG-34) and the USS Long Beach (CGN-9), and executive officer on the USS Jou- ett (CG-29).
[1]J.H. Parry, The Age of Reconnaissance (New York: Mentor, 1963), pp. 110-115. ‘A. Teixera da Mota, “Prince Henry of Portugal and the Progress of Nautical Cartography.” International Hydrographic Review, Vol. XXXIX. No. 1, January 1962, p. 36. . .
I0J.H. Parry, The Discovery of the Sea (Berkeley: University of California Press, 1974, 1981), p.149. .
"E.G.R. Taylor, The Haven-Finding Art: A History of Navigation from Odysseus to Captain Cook (London: Hollis and Carter, 1958), p. 162.
‘'Morison, Admiral of the Ocean Sea, Vol. I, p. 87.
’’Parry, The Discovery of the Sea, p. 190.
“Parry, The Age of Reconnaissance, p.l 15.
“Christopher L. Bewail, Portugal and the Quest for the Indies (New York: Barnes and Noble, 1974), pp. 215-216.
“Bell, Portugal and the Quest for the Indies, p.47. European mariners thus owed a great deal to Arab astronomers, which paradoxically may not have been true of the Arabs who sailed the Indian Ocean. Their astronomy had prelslamic roots, and as a tightly-knit, semi-literate society, they may have had no contact with the learned scientists of the caliphate. See J.F.P. Hopkins, “Geographical and Navigational Literature,” in Religion, Learning and Science in the Abbasid Period,
'Samuel Eliot Morison, Christopher Columbus, Mariner (Boston: Little, Brown and Company, 1955), pp. 198-199. Morison later argued that Ferdinand Magellan was the greatest navigator in history. See The European Discovery of America: The Southern Voyages, A.D. 1492-1616 (New York: Oxford University Press, 1974), p. 320.
'Clarence H. Haring, Trade and Navigation Between Spain and the Indies (Gloucester, Massachusetts: Peter Smith, 1918), p. 305.
’Morison, Admiral of the Ocean Sea: A Life of Christopher Columbus, 2 Vols. (Boston: Little, Brown & Co., 1942), Vol. I, pp. 243-244.
’Erie Axelson, Congo to Cape: Early Portuguese Explorers, George Woodcock, ed. (New York: Barnes and Noble, 1973), p- 46.
’Edgar Prestage, The Portuguese Pioneers (New York: Barnes and Noble, 1933 reprinted 1967), p. 321.
‘Morison, “Columbus and Polaris Part II,” The American Neptune, Vol. 1, No. 2, April 1941, p. 136.
’Morison, The European Discovery of America: The Northern Voyages A.D. 500-1600 (New York: Oxford University Press, 1971), p. 141.