I think the Internet has finally broken the HISTSCI_HULK; he’s lying in the corner sobbing bitterly and mumbling wrong, wrong, wrong… like a broken record. What could have felled the mighty beast?
29 January was the anniversary of the birth (1611) and death (1687) of the Danzig astronomer Johannes Hevelius and numerous people, including myself, posted or reposted articles about him on the Internet. One of those articles was the 2018 article, The 17th-Century Astronomer Who Made the First Atlas of the Moon by Elizabeth Landau, with the lede Johannes Hevelius drew some of the first maps of the moon, praised for their detail, from his homemade rooftop observatory in the Kingdom of Poland, in the Smithsonian Magazine.
I suppose that I’m really to blame because I let him read it. He was chugging along quite happily, nodding his head, and burbling to himself, on the lookout, as always, for history of science errors and howlers, when he let out a piercing scream, NOOOOOO!!!!!! And collapsed in a sobbing heap on the floor. I’ve tried everything but I haven’t been able to console the poor beast.
So, what was it that caused this total breakdown? The first six paragraphs of the article are harmless enough, with only some very minor questionable statements, not really worth bothering about, but then comes this monstrosity:
Mapping the moon was one of Hevelius’s first major undertakings. The seafaring nations at the time were desperately searching for a way to measure longitude at sea, and it was thought that the moon could provide a solution. The idea was that during a lunar eclipse, if sailors observed the shadow of the moon crossing a particular point on the surface at 3:03 p.m., but they knew that in another location, such as Paris, the same crossing would occur at 3:33 p.m., then they could calculate their degrees of longitude away from the known location of the city. More accurate lunar charts, however, would be required for the technique to be possible (and due to the practical matters of using a large telescope on a rolling ship, a truly reliable way to calculate longitude at sea would not be achieved until the invention of the marine chronometer).
One can only assume that it is an attempt to describe the lunar distance method for determining longitude but apart from the word moon, it has absolutely nothing in common with the actual lunar distance method. Put very mildly it is a complete travesty that should never have seen the light of day, let alone been published.
Lunar eclipses had already been used for many centuries to determine the longitude difference between two locations, but you don’t need either a map of the moon or a telescope to do so. Two observers, in their respective locations, merely record the local time of the beginning and/or the end of the eclipse (initial and final contacts) and the resulting time difference gives the difference in longitude. Lunar eclipses are impractical as a method of determining longitude for navigation, as they occur too infrequently; there will only be a total of 230 lunar eclipses in the whole of the twenty-first century, of which only eighty-five will be total lunar eclipses. For example, if you were sitting in the middle of the Atlantic Ocean on 6 June 2022 and wished to determine your longitude, you would have to wait until 8 November for the next total lunar eclipse. After that you would have to wait until 14 March 2025 for the next total lunar eclipse, although there are a couple of partial and penumbral eclipses in between.
Early Modern explorers did use solar and lunar eclipses combined with an ephemeris, a book of astronomical tables, to determine longitude on land, to establish their location and to draw maps. Columbus, famously, used his knowledge of the total lunar eclipse on 1 March 1504, taken from an ephemeris, to intimidate the natives on the island of Jamaica into continuing to feed his hungry stranded crew.
The lunar distance method of determining longitude is something completely different. It was first proposed by the Nürnberger mathematicus, Johannes Werner (1468–1522) in his Latin translation of Ptolemaeus’ Geographia, In Hoc Opere Haec Continentur Nova Translatio Primi Libri Geographicae Cl Ptolomaei, published in Nürnberg in 1514 and then discussed by Peter Apian (1495–1552) in his Cosmographicus liber, published in Landshut in 1524. For reasons that I will explain in a minute, it was found impractical, but was proposed again in 1634 by the French astronomer Jean-Baptiste Morin (1586–1656), but once again rejected as impractical.
The lunar distance method relies on determining the position of the Moon relative to a given set of reference stars, a unique constellation for every part of the Moon’s orbit. Then using a set of tables to determine the timing of a given constellation for a given fixed point. Having determined one’s local time, it is then possible to calculate the time difference and thus the longitude. Because it is pulled hither and thither by both the Sun and the Earth the Moon’s orbit is extremely erratic and not the smooth ellipse suggested by Kepler’s three laws of planetary motion. This led to the realisation that compiling the tables to the necessary accuracy was beyond the capabilities of those sixteenth century astronomers and their comparatively primitive instruments, hence the method had not been realised.
We now turn our attention to Landau’s closing statement in this horror paragraph:
More accurate lunar charts, however, would be required for the technique to be possible (and due to the practical matters of using a large telescope on a rolling ship, a truly reliable way to calculate longitude at sea would not be achieved until the invention of the marine chronometer).
Historically, tables of the necessary accuracy were produced by Tobias Meyer (1723–1762) in 1755. However, the calculations necessary to determine longitude having measured the lunar distance proved to be too complex and too time consuming for seamen and so Neville Maskelyne (1732–1811) produced the Nautical Almanac containing the results pre-calculated in the form of tables and published for the first time in 1766. One does not need a telescope to make the necessary observations. A sextant is sufficient to measure the distance between the moon and the reference stars and that had been invented by John Hadley (1682–1744) in 1731. The lunar distance method was in fact ready for practical use before the marine chronometer.
One question that I have, is did Landau extract this heap of nonsense out of her own posterior or is she paraphrasing somebody else’s description? Throughout her article she gives links to various books with the information she is using, so did she take this abomination from another source? If so, it is still out there somewhere creating confusion for anybody unlucky enough to read it. On the question of sources, Dava Sobel’s Longitude, which, despite her prejudices against it, contains a correct description of the lunar distance method was published in 2005 and the much better Finding Longitude by Rebekah Higgitt and Richard Dunn was published in 2014, so there is no real excuse for Landau’s load of bovine manure in 2018.
I don’t know how many people have subscriptions to the Smithsonian Magazine, but it has over 300,000 followers on Twitter. If we look at the Wikipedia article on the Smithsonian Institutions it starts thus, “The Smithsonian Institution, or simply, the Smithsonian, is a group of museums and education and research centers, the largest such complex in the world, created by the U.S. government for the increase and diffusion of knowledge (my emphasis), so why is the Smithsonian Magazine diffusing crap?
I’m hoping that with plenty of sweet tea and digestive biscuits, I’ll be able to restore the HISTSCI_HULK to his normal boisterous self.