Kelley was born here in Albany, New York. He had a long and distinguished career as an archaeologist and epigrapher, working to decipher the Maya hieroglyphs. He also was a diffusionist who believed that Barry Fell had on occasion correctly identified Ogham Irish writing in the New World, using it as evidence for what he described in 1990 as an “extensive European presence in the New World” in ancient times.
Kelley died in 2011.
Since I am not a subscriber to Pre-Columbiana, I can only go by what MacIsaac has presented in her article.
According to MacIsaac, in his posthumous article, Kelley compared the Maya calendar with the Chinese zodiac. This comparison is actually quite old, and it dates back to Alexander von Humboldt, from whose work Kelley’s comparison appear to be lifted. Both Humboldt and Kelley, for example, draw comparisons between the same set of calendars: Mexican, Chinese, Japanese, and Indian. Both men similarly identify the names of the days and their animal signs across these calendars. Both men also draw the same conclusion, namely that all of these calendars are corruptions of one ancient and perfect zodiac. So popular was this view that no less of an authority than the Encyclopedia Britannica claimed, for decades in the nineteenth century, that Mexico had been an outpost of Chinese civilization. Among many other reasons, they cited this:
Humboldt has shown that the Mexican calendar is identical in its principles, which are very artificial and complicated, with that which was in use among the Chinese, Japanese, Thibetians, Hindoos, and Tartars; and he has rendered it probable that the names of their days are borrowed from an extinct zodiac of 27 or 28 houses, anciently familiar to the same nations. (7th ed., s.v. “America”)
But by the end of the nineteenth century scholars had concluded that Humboldt was full of it, and he had correlated only the most superficial aspects of the calendars, namely the animals, finding connections between birds, mammals, and reptiles that were more likely coincidental. Historian Edward John Payne explained the problem in his unfinished volume on the Western Hemisphere, A History of the New World Called America (1892-1899), where he explained in detail (though with forgivable errors based on knowledge of the time) that the mechanics of each system were fundamentally different:
The reader is now in a position to judge for himself whether the Chinese and Mexican calendars, regarded in their substance, the method in which the days are coördinated with the longer natural units of time, or incorporated into artificial ones, have such a resemblance as to justify the inference that the one has been derived from the other, or indeed any resemblance whatever. The Chinese calendar is not merely lunisolar, but exhibits the lunisolar reckoning in its most advanced form; the Mexican takes no account whatever of the moon, and shows no trace of the most rudimentary attempt at coördination of the day with the lunation, or of the lunation with the year. In dividing the year into smaller periods it proceeds on a purely arithmetical principle, by the vicenary system; the arithmetic of China, so far as is known, has never been other than denary. The Chinese calendar presumes a knowledge of the year as a period of 365¼ days nearly; and this figure is incorporated as a fundamental element in the reckoning. The Mexican year, if our conclusions are right, was an uncorrected cycle of 365 days perpetually. The Mexican reckoning is in substance based on the number 13 as its corner-stone. Neither this, nor the other numbers (4 and 20) employed in the Mexican calculation of time, enters into the Chinese calendar in any way; conversely, no number used in the Chinese calculation of time enters into the Mexican calendar. The Chinese have at different times and for different purposes employed noctidiurnal cycles of 7, 10, 12, 28, and 60 days, and calculated periods of 12, 28, 60, 76, and 80 years; all these numbers are foreign to the reckoning of Mexico. The Chinese have from early times possessed considerable knowledge of astronomy, and have applied it to an accurate division of the noctidiurnal period of 24 hours into 12 equal parts. The Mexicans were wholly ignorant of the rudiments of astronomy, and possessed no means whatever of dividing the day into equal parts. The use of the technical term ‘cycle of years’ in comparing the Mexican period of 52 years with the Chinese period of 60 years, or any period founded on a coördination of the motions of the moon with those of the sun, is in itself misleading. Nothing was coördinated in the Mexican cycle of 52 years but two reckonings of days—two concurrent noctidiurnal cycles, each founded on the arithmetical process of simple multiplication, though each approximated, as nearly as current knowledge and modes of reckoning permitted, to a natural period. Such an arrangement is manifestly not a ‘cycle of years,’ in the sense given to that term by chronologists.
This fundamental fallacy affects the whole argument of Humboldt. The supposed resemblances on which his contention for imported advancement is based are wholly derived from the external aspect of the two calendars as two ‘cycles of years’ compared with each other; they are, in fact, neither more nor less than that in each case the ‘cycle of years’ is computed by two series of names or symbols, combined in the way above indicated, and that in each case some of the names used are those of animals.
While David H. Kelley more or less modernized Humboldt, a linguist from Tokyo named David B. Kelley tried to defend the other Kelley’s work in the same issue of Pre-Columbiana by using a computer program to correlate the Mexican and Chinese calendars in order to determine how often they would name the same animal on the same day. He found that when comparing the calendars from the Maya start date of August 11, 3114 BCE, the same animal names the day nine times out of sixty, which by my calculation would be 15% of the time, which does not seem statistically significant given the limited range of animals used to designate calendrical events in both cultures. Kelley was able to raise the correlation to 30 out of 60 by shifting the “start date” for the calendar to a random figure of no significance to Chinese or Maya cosmology.
The complexities of this correlation aren’t really worth getting into, but there are 20 days in a Mexican cycle, building into larger units in a 52-year cycle, while the Chinese calendar of the contemporary Han dynasty had 10-day weeks combined into 30-day cycles in a 60-year period. The Mexican calendars had names (animal, mineral, vegetable, or conceptual) for the days and the years, while the Chinese calendar names only years for animals. There are more than 100 variant Chinese calendars, so I am generalizing greatly, and since the Shang dynasty the ten-day week had for names the Ten Heavenly Stems (which are not animal names) that became astrological terms in the Han dynasty, according to vol. 3 of Joseph Needham’s Science and Civilisation in Ancient China (1959) and other sources. As Payne had written, the two calendars just don’t line up, their similarity being primarily that both cultures used a sort of “cog-wheel” calendar of interlocking cycles, albeit totally different.
The other Kelley had alleged, following Humboldt, that East Asian and Mesoamerican calendar animals derived from a “prototypical list” because of their similarities. Rabbits and monkeys were easy enough to correlate, but he dug deeper to prove his assumptions where they did not match clearly. He correlated the Mexican crocodile with the Chinese dragon. When one group in Guatemala had a turtle in its nineteenth position, but the Chinese did not, he still scored it a hit because the Malay people listed the turtle in the nineteenth position. The Aztec and most Maya did not have the turtle there but rather a lightning goddess, while the Hindus of India have a dog. This is all still a hit because an Aztec myth of no certain date says that this goddess was turned into a dog, and Buddhists depicted on one manuscript a dog sitting on a turtle. He says that the “lightning dog is found in Asia throughout the areas of Buddhist influence and is also found in Mexico.” Thus dog, turtle, and lightning must be the same because of a story unattested outside of central Mexico.
I wasn’t able to find any reference to a Buddhist “lightning dog,” which Kelley claimed was common. I found a few references to “thunder dogs,” the sidekicks of various storm gods, who personify thunder and lightning, but they seem to be an Austronesian concept. But even if it were common, so what? Pre-Christian European mythologies gave the Wild Huntsman a set of barking thunder-dogs who precede his Wild Hunt. The people of the Lower Congo also depicted lightning as a dog, and an Afro-Cuban folk tale originating in the Congo has a story of a turtle (or tortoise) pulling a fast one on a dog. Does that mean that the Congolese peoples are heirs to the same inheritance?
David H. Kelley and David B. Kelley both argued that Mesoamerican linguistic terms were so similar to Chinese as to be “interchangeable.” That claim I find hard to believe, given that Eugène Beauvois argued that Mesoamerican linguistic terms were so similar to those of Latin and Gaelic that they could only be adaptations of the same. Some of these terms would seem to be the same calendrical terms.
David H. Kelley concluded that the Maya adopted the Chinese calendar around 200 CE, which would be a surprise to the Mesoamericans who had been using their own calendrical system since at least 500 BCE, according to modern scholars.