Christopher Columbus – Caribbean Archaeology Program

Where did Columbus really set foot on the New World? Theories and sites abound.

by William F. Keegan

Originally published in VISTA magazine, October 6, 1991.

To read Columbus’s daily log (diario de a bordo) you would think that his small fleet was never very far from land. For 32 days after leaving Gomera in the Canary Islands on September 9th, the diario makes repeated reference to signs of land. Sailing in the middle of the Atlantic Ocean, more than 1,000 miles from the nearest land, Columbus observed “river weed” (sargassum seaweed), a live crab “not found more than 80 leagues (240 miles) from land,” a booby or gannet, birds that “do not depart more than 20 leagues from land,” and “a large cloud mass, which is a sign of being near land.” But it was not until two hours after midnight, the 12th of October, that land finally did appear.

The land was an island, which the native Lucayans called Guanahani, and Columbus renamed San Salvador (“Holy Savior”). Scholars agree that Guanahani is in the Bahama archipelago, but that is where agreement ends. To date, ten different islands have been identified as the first landfall; a truly remarkable number when you consider that only 20 islands in the entire archipelago are even remotely possible candidates. In addition, more than 25 routes have been proposed to take Columbus to the three other Lucayan islands he visited before departing for Cuba. Represented on a single map these routes look like someone gone mad playing connect the dots.

Cat Island, in 1625, was the first to be proposed as the landfall island. Cat went unopposed until Watling Island was suggested in 1793. Grand Turk was next, followed by Mayaguana, and Samana Cay in time for the 400th anniversary in 1892. Cat Island’s claim was ably defended by the novelist Washington Irving (“The Legend of Sleepy Hollow”), while Watling was promoted by the Chicago Herald (site of the Columbian Exposition in 1893), and Samana was championed by Gustavus Fox who served as Assistant Secretary of the Navy under President Abraham Lincoln.

In 1926, Cat and Watling entered a legal battle over who had the right to use the name San Salvador. The case was settled by the Bahamas legislature in favor of Watling. Known legally as San Salvador ever since, Watling gained its strongest support from the distinguished Harvard Historian Samuel Eliot Morison who retraced Columbus’s steps in his 1942 Pulitzer Prize winning biography of Columbus. Morison’s reconstruction seemed to end the debate once and for all.

Other first landfall islands have been suggested since — Conception (1943), East Caicos (1947), Plana Cays (1974), Egg/Royal (1981), Great Harbour Cay (1990) — but none has made a sufficiently strong case to sway popular opinion away from Watling. None, that is, until 1986 when National Geographic magazine told 40 million readers that Samana Cay was the place.

But why the debate? Why hasn’t Guanahani been identified with certainty? The answers lay in the quality of the evidence. The only detailed information concerning Columbus’s first voyage is contained in his diario. Columbus presented the original to Queen Isabel who had a copy made for Columbus. The whereabouts of the original are unknown, and all trace of the copy disappeared in 1545. What has survived is a copy made by Bartolomé de las Casas — a thirdhand manuscript handwritten in sixteenth-century Spanish that has numerous erasures, unusual spellings, brief illegible passages, and notes in the margins. The ambiguities, errors, and omissions in this manuscript have been compounded in modern-language translations.

Putting such problems aside for the moment, what of that account might be used to identify Guanahani? Arne Molander, an advocate of Egg/Royal Island, has identified 99 clues, many of which require specialized knowledge and most of which are subject to multiple interpretations. Such minutia are beyond the scope of this brief article, instead let us consider four general categories: ocean crossing, descriptions of the islands, sailing directions and distances, and cultural evidence.

Using a computer generated simulation of the first voyage that took into account prevailing winds and currents, the National Geographic team concluded that the crossing ended at Samana Cay (actually, they overshot Samana by more than 300 miles and had to shorten their league by 10% to land at Samana). When a team from Woods Hole Oceanographic Institution substituted average for prevailing winds and currents, their simulated crossing ended in sight of San Salvador (without need to adjust for distance). However, not satisfied with that solution, this same team plugged new numbers into their computer and put Columbus near Grand Turk! Too bad, as one reviewer noted, Columbus didn’t have a computer on board.

A different approach to the crossing is to simply use Columbus’s statement that Guanahani was on the latitude of Ferro in the Canary Islands. Simple enough? Latitude sailing was certainly possible in Columbus’s day, and Arne Molander has shown that the latitude from Ferro crosses Egg Island, just north of Eleuthera. However, Robert Power, armed with maps of the day, has shown that the Americas are consistently displaced northerward on these maps and that in sixteenth-century cartography the line from Ferro crosses Grand Turk. In this way both northern and southern Bahamas landfalls have been supported.

The situation does not improve when you move to descriptions of the islands themselves. For example, prospective Guanahanis range in size from 10 to 389 square kilometers, the harbor that could hold “all the ships in Christendom” from .6 to 36.6 square kilometers, and the second island is either 5 by 10 leagues (as recorded in the diario) or 5 by 10 miles (a likely transcription error).

If we cannot be certain what he was describing, then we should at least be able to retrace how he got there. Yet the record of directions and distances has been used to defend more than 25 different routes. The most basic disagreements concern translation; such as whether camino de should be translated as “the way from” or “the way to.” More complicated disagreements arise over interpolations. Between the night of October 17th and the morning of the 19th one route has the fleet sail fewer than 20 miles, while another has them cover more than 300. The first claims that bad weather prevented them from sailing on the 18th while the latter claims that storm winds propelled the three ships at breakneck speed.

Lastly, Columbus visited four native villages and spent three days trying to reach the village of a chief. I have used archaeological evidence to show that the Watling to Rum Cay to Long Island to Crooked Island to Cuba route best fits all of the data. Others, however, believe that there were so many Lucayans living in the Bahamas that virtually every route will find archaeological sites in the places where Columbus observed villages. Only more archaeology will tell.

Where was Columbus’s first landfall in the Americas? The Lucayans called the island Guanahani, and Columbus renamed it San Salvador. In my opinion it is known today by the name Columbus gave it.