1421

Steven Dutch, Natural and Applied Sciences, University of Wisconsin - Green Bay
First-time Visitors: Please visit Site Map and Disclaimer. Use "Back" to return here.


Between 1405 and 1421, the Chinese, for the only time in their history, engaged in long distance sea explorations, sending voyages across the Indian Ocean as far as East Africa. These voyages were the most massive sailing voyages ever undertaken, involving dozens of ships and thousands of crew members. After 1421 the Ming Dynasty sharply curtailed sea travel (some say abolished it altogether) and, in an intellectual crime comparable to the Spanish destruction of Maya literature or the U.S. abandonment of the Apollo Program, destroyed most of the records of the voyages.

The destruction of the sailing records means that much about the voyages remains conjectural. It is by no means impossible that the Chinese rounded the southern end of Africa and saw the Atlantic. It seems impossible that they would not have known of Australia. Asian vessels were extremely seaworthy and chance voyages by fishing or trading vessels across the Pacific are certainly possible. Gavin Menzies, a former British Royal Navy officer, has an interpretation that puts all these ideas in the shade. In 1421: The Year China Discovered America, he argues that the last great Ming voyage was a global exploration venture that took the Chinese to every continent except, curiously, Europe.

If anyone could have done it, the Chinese could. But did they?

Anyone familiar with the works of Immanuel Velikovsky, Charles Berlitz or Erich von Daniken will feel right at home with Gavin Menzies. The flavor is exactly the same. It's all there: Piri Reis map, check; Bimini Wall, check; unexplained ruins, check; long lists of alleged wonders with no documentation, check; prosaic illustrations taking up space instead of actual pictures of alleged finds, check. The only difference is this time it's not aliens or Atlanteans, but Chinese who are responsible.

Page references are to 1421: The Year China Discovered America (Harper Perennial edition, 2004).

What Menzies Doesn't Have

Navigational Howlers

The Chinese could not determine their latitude south of the equator, where Polaris was invisible. It was a problem that had to be solved. A star or stars in the southern hemisphere that could fulfill the function of Polaris in the northern had to be identified before Zhu Di's dream of charting the whole world could be realized (p. 92)

For a career naval officer to make such a statement is unbelievable - it reveals a complete ignorance of celestial navigation. First, there's nothing magical about Polaris - it's simply an accident that right now there is a bright star close to the north celestial pole. Precession (a term not in the index of Menzies' book, by the way, though he mentions it a couple of times) will bring the north celestial pole closer to Polaris for about another century. In 1421, Polaris was over three degrees from the north celestial pole. It was still the best north pole star, but would have moved very perceptibly in the course of a night. If you naively assumed Polaris marked the north celestial pole, your determination of latitude could have been off by up to 360 kilometers . If you assumed Polaris marked due north, your course headings would have varied as much as 6 degrees in a single night.

Second, the search for a corresponding star in the southern hemisphere will be a very long and fruitless one - or a very short one. Right now the south celestial pole is in a pretty barren patch of sky, and there is no conspicuous southern pole star, nor has there been one, nor will there be one, for many centuries. And it wouldn't require a long search to determine this - a single night of observation would confirm it.

The reality is you don't need a pole star at all to determine your latitude. If you know the position of any star relative to the celestial poles (and the Chinese did to good accuracy) then you can measure its elevation when the star is at its highest. Your latitude is 180 - (Altitude of Star) - (Star's distance from the celestial pole).

How do you determine when the star is at its highest? Well, that's why ships have people on watch. Anyone decently familiar with the constellations can locate directions to pretty good accuracy, so it would not be hard to tell which stars are close to due south or due north. Periodic measurements would identify the exact moment and altitude, a measurement that would also provide the direction of the meridian.

Finally, there was a nifty gadget called an astrolabe (another term not in Menzies' index) that was well known to the Arabs and even in Europe by 1421. An astrolabe consisted of a stylized star map, called a rete, and a series of elevation and direction grids called maters, usually consisting of a set of maters for different latitudes. By measuring the altitudes of three stars and positioning the rete on different maters for best fit, it would be possible to determine both latitude and direction.

You don't even need an astrolabe to use this method. Menzies himself provides a Chinese example (p. 90). It's an excerpt from Chinese sailing directions to the Indian Ocean, and it refers to three stars or asterisms in different directions and their elevations above the horizon. There's no instruction on how to find the directions since any captain who couldn't tell approximate directions at sea by a glance at the stars wouldn't be qualified to pilot a rubber duck in a wading pool, much less a massive junk across an ocean. The stars are two stars in Orion setting in the west and the triangle of Lyra rising in the northeast. These are obviously useful only at certain seasons. They wouldn't be useful in May and June when the sun is too close to Orion, but if you're sailing with the monsoons, that limitation wouldn't matter.

Menzies has a lengthy appendix of evidence to back up his claims, virtually all of it so poorly cited that it's impossible to check. But it includes a couple of dazzling navigational boners:

Due to the earth's precession, the North Pole as determined by Polaris at 90o elevation was approx. 200 nm nearer to Greenland than it is today (p. 569)

No, the North Pole is determined by the rotation of the heavens (speaking geocentrically), not the elevation of Polaris. The North Pole in 1421 was exactly where it is today.

Chinese circumpolar star charts, which could only have been drawn by people who had seen the stars at a latitude north of 73 degrees (p. 569)

I hardly know where to start with this one. The closer to the equator you are, the more of the sky you can see, so every star visible at 73 north is visible at lower latitudes in the northern hemisphere. And once you understand what "circumpolar" means - stars that never set - you can draw a chart of circumpolar stars for any latitude whatsoever given only an accurate star chart. Circumpolar stars are those closer to the north celestial pole than your latitude; at 30 north, for example, the circumpolar stars are in a circle of 30 degrees radius centered on the north celestial pole. Needless to say Menzies doesn't reproduce this chart.

There's another problem. At 73 north, the sun is continuously above the horizon from early May to early August, and stars won't become easily visible until the sun gets 10 degrees or so below the horizon. For nearly half the year - the time most likely to have open water - the stars won't be visible at 73 north. The only time they would be easily visible would be from September to April, when conditions would be too harsh for navigation.

The Piri Reis Map

Even by the standards of most of the nonsense published about the Piri Reis Map, Menzies' interpretation stands out.

A simplified version of the map is shown at left. The red radiating lines are compass bearings. Charts of this type, called portolan charts, were standard in the Middle Ages and Renaissance. Africa and Europe are quite recognizable.

The northernmost horizontal line on the map is a very good approximation for 45 degrees north and the line just below the bulge of Africa in the center of the map is a good match for the equator.

The horizontal line through the northern large circle is ambiguous. It might represent the Tropic of Cancer, but since it's exactly between the equator and 45 north, it might also represent 22.5 north (that is, 1/16 of a circle).

The fact that the western land mass looks like Brazil and the equator cuts the coast only a bit too far north pretty conclusively identifies that bulge as Brazil. Not to Menzies.

Here's how Menzies matches the map to South America, in red (p. 154). The Piri Reis coastline is on the left and a modern map of South America is inset to the right.

Menzies completely ignores the latitude indications on the Piri Reis map and matches features on the Piri Reis map with features 45 degrees to the south.

Then he completely ignores scale. He matches 1100 kilometers of Patagonian coastline with a coastline on the Piri Reis map that is bigger than the bulge of Africa (3400 kilometers!)

Actually, Menzies describes his methodology (p. 155) - he began by assuming the map shows Patagonia, then determining scale from that assumption.

Where did the rest of South America go, in Menzies' interpretation? Surely the bulge of Brazil would have been a lot better surveyed by the early 1500's than Patagonia.

After all this, completely omitting Tierra del Fuego is but a minor quibble. Presumably mariners might be interested in knowing whether a coastline faced the open sea or was one side of a strait, but, hey, I wasn't a career naval officer in the Royal Navy. What do I know?

In case you still are interested, the coastline to the south is supposedly actually sea ice connecting to Antarctica. So we had ice reaching all the way to South America (55 south). At the same time, the Chinese supposedly circumnavigated Greenland, reaching 84 north in open water.

The Kangnido Map (1402)

Menzies writes a lot about the Kangnido map of 1402, a Korean world map, and although he shows details of the western part of the map, showing Africa and Europe, nowhere does he reproduce the entire map. So here it is.

We can see several things at a glance.

Africa's Missing Bulge

The western part of the Kangnido Map is shown at left. The recognizability of the Mediterranean is really remarkable, considering how many degrees removed the mapmaker was from the original sources of data. Spain, Italy and Greece are all clearly recognizable, as is Sicily.

On the other hand, Arabia is crude and India and Madagascar are missing. The Nile is shown emptying into the Red Sea. The Horn of Africa is missing.

Actually, the Mediterranean looks a lot more like some plate tectonic reconstructions of the Late Mesozoic, what with the Bay of Biscay being closed and Italy and Greece as islands. If you want to argue the Chinese knew about plate tectonics in the 1400's, be my guest. I get the royalties.

Menzies deals with Africa's bulge by invoking the South Equatorial Current, which he claims carried the Chinese fleet westward without their being aware of it. Since they were in a current, their speed relative to the water would have been small or zero, and they wouldn't have realized they were moving. Therefore, he argues, they failed to realize that the coast swung far to the west (below).

Pseudoscience and the "Practical" Thinker

Menzies' description of the currents is faulty at best. The South Equatorial Current doesn't "peter out a thousand miles into the Atlantic" (p. 130) - it forms the northern part of the South Atlantic Gyre. Furthermore, along the equator is a narrow east-flowing current, the Equatorial Countercurrent. Equatorial countercurrents occur in the Atlantic, Pacific, and Indian Oceans, and are a consequence of the Coriolis effect. When water hits the west side of an ocean basin, it has to go somewhere. Some flows north and some flows south. Water that flows away from the equator (most of it) becomes part of one of the great oceanic gyres. Any water that flows toward the equator can only flow back east, and once it does, the Coriolis effect will force it to remain along the equator. In the Bight of Benin, the interaction of the Equatorial Countercurrent and the South Equatorial Current creates a small loop. Thus, it's just as likely that currents in this region would have carried the Chinese east instead of west, or carried them in a circle.

This digression illustrates an important aspect of pseudoscience. Whenever we find someone with technical training espousing pseudoscience, odds are strong that the person will be an engineer, a military officer, an astronaut, or some other "practical" type. Like the engineers who espoused Velikovsky, astronauts looking for Noah's Ark on Mount Ararat, or Menzies harking back to his naval experiences for credibility, they tend to assume that their "practical" experience makes them superior to "ivory tower, theoretical" types.

The reality is exactly the opposite. "Practical" people so often espouse pseudoscience precisely because they have poor theoretical understanding. For all his years in the Royal Navy, Menzies does not know celestial navigation, nor does he know the oceans, nor does he really know very much about maps. The reason that some engineers championed Velikovsky while others have backed creationism is that they do not know enough science to understand why Velikovsky and creationism are wrong. Faulty science, in turn, can lead all too often to bad practice. After the Challenger disaster of 1986, engineers on the inquiry panel complained they couldn't see how to test the idea that cold temperatures might have made rubber gaskets of the Challenger's rocket boosters stiff. Physicist Richard Feynmann, as "ivory tower" and theoretical as they come, calmly dipped a sample of the gasket material into a glass of ice water.

When a "practical" person espouses pseudoscience, his "practical" knowledge doesn't elevate the theory to respectable. Instead, it shows that his understanding of science is so poor he's probably not all that good at his "practical" calling.

Africa and the Kangnido Map

Above is a real map of Africa compared to the Kangnido Map. Far from being a fabulously accurate rendition of Africa, the Kangnido Map bears no resemblance to the east coast of Africa.

The Vinland Map

Everyone surprised that Menzies accepts the Vinland Map (left) as authentic, raise your hand.

The Vinland Map, purportedly made about 1450, has been controversial ever since it first surfaced in 1957. Its original provenance is unknown. The most widely reproduced part of the map is at left, but there's another half showing Russia and Asia

The most widely cited evidence that the map is a forgery is that the ink contains a form of titanium dioxide called anatase, an ingredient not used in old inks. However, since anatase is a naturally occurring mineral, its presence in the ink is not as conclusive as Vinland Map detractors believe.

Anatase in fine powder form is white and used in inks mostly for opacity. It's also quite rare in nature. Its rarity and the difficulty of grinding things by hand to the size of the ink particles in the map are probably the most suggestive arguments for modern origin. Still, one can see a medieval scribe, ignorant of chemistry, mineralogy and the physics of light scattering, grinding up a promising mineral, finding it helped make a better ink, and using it without caring much what was in it or why it worked.

The map could easily be a forgery - the cartography fits what people would like to see too neatly - but the anatase argument is not quite as air-tight as map critics often insist. The fact that anatase occurs in nature, even if it is rare, means its presence in the ink cannot be conclusive evidence of forgery.

Menzies likes the accurate portrayal of Greenland, but dismisses the obvious source of information - the Vikings - in favor of the Chinese.

The Waldseemuller Map (1507)

At left is North America as shown on the Walseemuller Map of 1507, the first map to use the name "America." According to Menzies (p. 464):

"The Waldseemuller (1507) shows nearly a thousand miles of the Mississippi as well as the Brazos, Alabama, Roanoke, Delaware and Hudson Rivers of North America."

Identifying those rivers is left as an exercise for the reader, as well as explaining how Menzies knows the "Delaware" isn't, say, the Susquehanna or Potomac, or the "Brazos" isn't, say, the Trinity, Sabine, or Rio Grande, or the "Hudson" isn't the Connecticut or the Penobscot.

It might be tempting to dismiss the distortion of North America as an artifact of the projection, but the meridians don't allow it. In addition, Waldseemuller also published gores for a globe (below), and the Americas are shown in the three rightmost gores. It's obvious that Waldseemuller really pictured the Americas as narrow. On the large map and the globe, he shows a narrow gap between North and South America, but on an inset in the large map he shows a connection.

By interesting coincidence, he has a meridian close to our present zero longitude, and he puts the tip of Asia 220 degrees east of that, or 140 degrees west longitude, barely 15 degrees west of the present coast of Oregon. Clearly he was clinging, or catering to, the hope of a short sea passage west to Asia. His estimate of the longitude of the tip of Brazil is right on the money, but he has North America much too far west.

The Jean Rotz Map (1542)

Unlike many early maps, this one scrupulously shows only what is known. Many areas that could have been filled in given the knowledge of the time are not, for example, southern Greenland, northern Scandinavia, and most of China. The full map is shown above: the area of interest with the Renaissance ornate decoration cropped out, is shown below.

What leaps out from this map is that it contains nothing that a European wouldn't have known in 1542! Areas well explored, like the Mediterranean and West Africa, are fairly accurate. Areas more recently explored, like the rest of Africa, India and Southeast Asia, and the Caribbean, are quite recognizable but have noticeable errors. India is too narrow and the Horn of Africa much too small. The Isthmus of Panama is shown crudely but recognizably for the very good reason that it had been crossed by Europeans. The Pacific coast of South America is tolerably accurate down to northern Peru because the Spanish were by this time invading Peru.

On the other hand, what are we to make of eastern Brazil being an island?

Menzies is most enthralled with the large land mass roughly corresponding to Australia, although the actual coastlines bear virtually no resemblance to Australia. The far eastern end with the two large bays he identifies with Auckland and Campbell islands. The intervening coastline? Sea ice. At 50 degrees south. This at a time when he simultaneously asserts that Greenland (reaching almost 84 north) was circumnavigable and surrounded by ice free waters.

Nowadays, the limit of Antarctic sea ice is around 60 degrees south in the southern winter, retreating in many cases to the coast of Antarctica during the summer. It never gets anywhere near Campbell or Auckland Islands and Menzies offers no real evidence that it did in 1421, either. Nor does he explain why skilled Chinese navigators would risk their ships in ice-filled seas during the southern winter.

Actually, you can make a far better case that the far eastern end of the map shows New Zealand. The latitude is off by ten degrees, but the orientation of the coast, the sharp northern end and the prominent channel a third of the way down are a far better fit to New Zealand than the Kangnido Map is to Africa.

Rule #1 For Interpreting Ancient Maps (If You Want A Best Seller)

Anything that matches (or can be made to seem like a match to) existing cartography is proof that the cartographer had access to secret knowledge. Anything that doesn't match, doesn't count.

Menzies and the Little Ice Age

With all the debate about global warming, you'd think Menzies would have loads of documentation about Arctic climate. Wouldn't the Little Ice Age have affected the Chinese ability to circumnavigate Greenland, for example?

The term "Little Ice Age" is confusing because it has been so variously used. The most detailed climate records suggest cooling from about 1300 to 1800, with a respite in the 1500's. Some authors use the term to refer to the earlier event, some for the latter, some for the entire interval. Also, attempts to reconstruct global temperatures don't show much of a dip, however well climate indicators document a real cooling, so the Little Ice Age was apparently primarily a northern Hemisphere event. That doesn't exactly mesh with Menzies' postulates of sea ice extending to Tierra del Fuego and nearly to Tasmania, while simultaneously allowing the Chinese to sail in open water around Greenland, if not to the North Pole itself.

Instead, Menzies simply ignores the literature on the Little Ice Age and asserts "The 'mini Ice Age' started in 1432" without citing a single published source (p. 569). The tem "Little Ice Age" doesn't even occur in the index at all.

Code Talkers

Menzies asserts (p. 559):

Really makes you wonder how the famous code talkers of World War II pulled it off. If there's that much mutual intelligibility between Southwest Indian and East Asian languages, why didn't the Japanese break the code easily? (Yes, intelligibility isn't always bilateral; Spanish speakers can often understand Italian speakers, but not vice versa; but if I tried to frustrate Spanish listeners by speaking Italian, it wouldn't take long for the listeners to break the code.)

The Maya in Wisconsin?

According to Menzies:

Mayan miners and astronomers knew of copper deposits in northern Michigan and sent expeditions to establish control of the area. They built a large settlement at 'Aztalan,' which became the center of copper trading for several centuries, but in about 1300, copper deposits were found in Mexico and Azralan was apparently abandoned.

...Taken altogether, all this evidence makes it at least arguable that the Chinese came to mine copper at Azrtalan.

Rarely do you see so much garbled information in so few lines. Aztalan is a state park in southern Wisconsin that preserves the unfortunately damaged remains of a site that was occupied about 900-1200 AD. Actually Aztalan was named by 19th century settlers for the mythical homeland of the Aztecs. It was one of the northernmost communities of the Mississippian mound culture, whose largest settlement was at Cahokia in Illinois, and the evidence for a Maya connection to Aztalan is precisely zero. And why would anyone build a settlement 200 miles from the ores they were supposedly trying to mine?

At left is one of the large mounds at Aztalan, together with a reconstruction of its palisade.

Note how easily we slide from the Maya at Aztalan to the Chinese. Now to get from Aztalan to the copper deposits in northern Michigan, you pass over some very rich iron ore deposits. The Maya wouldn't have known they were valuable, or how to exploit them, but the Chinese most definitely would. So why didn't they? Pre-Columbian iron smelting in North America would be a revolutionary discovery - where's the evidence?


Return to Pseudoscience Index
Return to Professor Dutch's Home Page

Created 18 August, 2006;  Last Update 02 June, 2010

Not an official UW Green Bay site