SIAM News Blog

How Did Newton Go So Far Off Course?

By Ernest Davis

BOOK REVIEW: Newton and the Origin of Civilization. By Jed Z. Buchwald and Mordechai Feingold, Princeton University Press, Princeton, New Jersey, 2012, 544 pages, $49.50

The sun (black circle) setting in southern Grecian latitudes on the day of the vernal equinox in 939 BC, the year, according to Newton, in which Chiron delineated the first celestial sphere—one of several arguments in Newton’s Chronology that proved to be inaccurate. From Newton and the Origin of Civilization.
Isaac Newton’s book The Chronology of Ancient Kingdoms Amended was published in 1728, a year after his death. Like much else in Newton’s career, the appearance of the book was mired in unedifying controversy. It was published largely at the urging of William Whiston, Newton’s successor as Lucasian Professor of Mathematics at Cambridge, once an acolyte, now a bitter enemy, in order that he could respond with a crushing refutation. Three years earlier, an unauthorized, abridged version of the book had been published in France, together with a short refutation; the aged Newton was, reasonably, furious.

In the course of the inevitable controversy over the pirated edition, Newton minimized the significance of his own work, saying that he “had occupied himself agreeably with history and chronology when fatigued with other work.” In fact, he had had a serious interest in chronology from a fairly early age, and since about 1700 had applied himself intensely to it. The subject constitutes a significant fraction of the immense body of manuscript notes that he left behind. He had compiled a large library of both primary and secondary sources. Besides Latin, he had a good knowledge of Greek and some knowledge of Hebrew. He analyzed the sources critically, taking into account reliable information that would have been available to the authors, and what the authors’ biases would have been. And, of course, Newton was a genius.

Nonetheless, as viewed with 20/20 hindsight after nearly 300 years, the Chronology made no useful contribution whatever to historical research, either factual or methodological. As far as I can determine, the Chronology has nothing new in it that is right, and it includes major errors that practically no one else, before or since, has made. Most egregiously, Newton claimed, on the flimsiest of grounds and in the face of massive contrary evidence, that pharaonic Egypt was not a kingdom of major significance until after the time of King Solomon. The Chronology is also extremely dull and inhuman; people live and die, cities are founded and destroyed, kingdoms rise and fall with no trace of individuality or character, only intricate argumentation about dates.

How did Newton go so far off course? To begin with, in the early 18th century, a scholar trying to determine reliably any date prior to about 600 or 700 BC faced formidable, perhaps insuperable, obstacles. Archeological evidence was limited to a few undatable monuments, like the Pyramids; the only evidence was textual. The surviving histories of the early period were written many centuries after the events they described; they combined history, myth, and sheer fiction; and they were often polemics or propaganda aimed at establishing the antiquity or early supremacy of whichever group the author favored. It was essentially impossible for Newton or his contemporaries to establish a clear history. Viscount Bolingbroke, a generation younger than Newton, worked avidly on ancient chronology for a while, and then gave up in despair, writing “Who can resolve to build with great cost and pain when he finds, how deep soever he digs, nothing but loose sand?” What is surprising is Newton’s extraordinary confidence; he claimed that all the dates in his history were accurate to within 10 or 20 years.

The argumentation in the Chronology is so intricate and so intertwined that it is hard to separate out particular errors, but a few stand out. Newton was a firm believer in the Euhemerist theory of mythology, which posits that mythological figures all derived from historical persons. This was a dangerous starting point for a chronologer, because it put a premium on the vain pursuit of dates for the originals of Hercules, Osiris, and so on. Much worse, Newton tried to identify corresponding historical and mythological figures—Saturn and Noah, Jupiter and Shem, for instance. In particular, the identification of the Egyptian pharaoh Sesostris with Osiris, Dionysus, and the Biblical figure Sesak placed a huge strain on the chronological structure.

Newton was quite skeptical of the written sources; Buchwald and Feingold argue that he became more so after, as Master of the Mint, he had to conduct investigations into cases of counterfeiting and to sift the lies that both culprits and witnesses told with abandon. But his skepticism deserted him at exactly the wrong times. He threw out as unreliable many ancient historical texts, such as the chronology of Egypt written by Manetho in (probably) the third century BC, but he was entirely confident of his own unfounded theory that all ancient civilizations derived from the Israelites.  He was willing to reinterpret some parts of the Biblical text; yet he took the flood of Noah at face value, and therefore set the dates of the creation of large kingdoms late enough to allow for sufficient population growth from the eight people who survived the flood.

Because texts and textual analysis were unreliable, Newton put his faith in a number of astronomical arguments. The most important for his system was this: The Greek astronomer Hipparchus quoted a passage from the earlier astronomer Eudoxus that lists the constellations that lie on the great circles connecting the poles with the equinoxes and with the solstices. Hipparchus found the list of Eudoxus incorrect. Newton conjectured that Eudoxus was in fact describing a celestial globe created by the much earlier astronomer Chiron, a contemporary of the Argonauts, and moreover that what Hipparchus had considered a mistake was entirely a result of the change in the position of the equinoxes and solstices due to precession in the centuries between Chiron and Hipparchus. A straightforward astronomical calculation makes it possible to determine the date at which the description in Eudoxus’ text would have been accurate. Newton concluded that the sphere was constructed around 939 BC and that the Argonauts’ voyage was about a generation earlier; the standard date given for the voyage in chronologies based on Greek historical texts was three hundred years earlier than that.

Serious weaknesses in the argument were gleefully picked apart by his opponents in the years following 1728. First, the text of Eudoxus states only that the great circles pass through the constellations, not where they pass through. Different placement of the great circles within the specified constellations would change the results of Newton’s calculations by centuries. Newton arbitrarily decided that the great circles went through particular stars, which he chose to make the dates come out the way he wanted. Second, the association of the hypothetical celestial globe with Chiron was pulled out of thin air.

Buchwald and Feingold analyze the whole of Newton’s chronological work in the context of his life, his scientific work, contemporary chronology, and contemporary thought. Their erudition is staggering, and their labors were Herculean. They have read all of Newton’s manuscripts on the subject, deciphering documents that Newton first wrote left to right, and then wrote over top to bottom, and that mix chronological notes with whatever else Newton was working on at the time. They have read and analyzed everything with the remotest bearing on the chronology. They seem to have reworked not only all the calculations that Newton and the other chronologers actually did, but all the calculations that they could have done instead. They even analyze the way that Newton dog-eared pages of the books in his library, illustrated with a photograph.

The book makes little accommodation to the non-expert reader. The ideal reader would be effortlessly familiar with the Chronology; with Newton’s biography; with all the sources, primary and secondary, that Newton used; and with the whole state of chronological science in Newton’s time. I am not that reader, by a long shot. I advise any reader whose knowledge falls short of this to start by reading the first major study of Newton’s Chronology, Isaac Newton: Historian, by Frank Manuel (1963), which is much less daunting.

Even for less learned readers, however, Buchwald and Feingold’s book is full of remarkable details, insights, and incidents. There are detailed discussions of Newton’s pioneering use of computing an average in order to combine unreliable data, as opposed to trying to decide on the best value; of contemporary theories of vision and knowledge; of debates about the best technique for measuring celestial angles; of the manufacture of telescopes and of celestial globes, both ancient and contemporary; of the way in which the pictures associated with the constellations were placed over the corresponding stars, a critical question when you need to interpret phrases like “the end of the jaw” and “the tip of the nostril” of the constellation Cetus, which occur in Ptolemy’s star catalogue; and much else. Certainly, the book is essential reading for anyone with a serious interest either in Newton or in the study of ancient history in the 17th and 18th centuries.

Very recently, Newton’s Chronology has been taken up by fundamentalist Christians, because it fits with the Biblical account and bears the imprimatur of the great scientist. (One review on of a recent reprint recommends it for home schooling.) My first thought, on seeing this, was that Newton would roll over in his grave. My second thought was that my first thought was historically unjustifiable, merely a projection of my own feelings onto a very strange man, with a very different viewpoint, who lived in a very different world, and who died almost 300 years ago.

Ernest Davis is a professor of computer science at the Courant Institute of Mathematical Sciences, NYU.

blog comments powered by Disqus