Hipparchus’s Star Catalog NOT discovered!

Since October 2022, the international press has been impressed by the fact that the star catalogue of Hipparchus of Nicaea has supposedly finally been found (reports even in Nature, Research and Knowledge, Daily Heritage and many more). As with previous claims of this kind, this once again turned out to be false (current JHA paper).

While so far mostly the Roman marble globe in the Villa Farnese (Naples) has been interpreted as presumed to be of Hipparchian origin (which was regularly refuted after the claim was made), there is now a new source: a medieval parchment, Codex Climaci Rescriptus (CCR), which bears a text in the extinct Syriac language, but under which an older text still shines through.

Multispectral photography enabled a team of researchers to read the erased text beneath the plain text in 2022. The older text that was ‘erased’ is not Syriac, but Greek. It is not written in the (later developed) cursive script, but in the older block script and is therefore likely to date from antiquity and not the Middle Ages. The ingenious thing is that the team led by Old Testament researcher Peter Williams (from the Faculty of Divinity at the University of Cambridge, UK) (a) discovered this text, (b) was able to read it using modern technology, and thus (c) recognised that it is an astronomical text. In June and December 2022, this British team published a paper and the associated data entitled

“Newly-Discovered Illustrated Texts of Aratus and Eratosthenes…” 
Paper with all sources from Dezember 2022 in “Classical Quarterly” of the Unversity of Cambridge.

So far, so right.

The Mistake for the History of Astronomy

In October 2022, however, a paper appeared in the Journal for the History of Astronomy (JHA) that claimed to be about

‘New evidence for Hipparchus’ Star Catalogue…’ (link)

Knowledgeable readers of Hipparchus will immediately suspect this to be not right. Nevertheless, the wild hypothesis received enormous public attention (see above).

In the abstract, the authors write

1. It also confirms that Ptolemy’s Star Catalogue was not based solely on data from Hipparchus’ Catalogue.
2. Finally, the available numerical evidence is consistent with an accuracy within 1° of the real stellar coordinates, which would make Hipparchus’ Catalogue significantly more accurate than his successor Claudius Ptolemy’s.

and further, in the main text:

1. …constellation in μῆκος (‘length’) and πλάτος (‘breadth’), … The concept of constellation boundaries underlying both sections is both similar to and different from its present-day analogue: these boundaries are drawn along vertical lines of right ascension and horizontal parallels, like today, but they make up simple rectangles instead of the intricate shapes introduced by Eugène Delporte (1882–1955).
2. These coordinates are accurate to within 1° for the epoch of Hipparchus’ star catalogue (ca. 129 BCE),¹¹ as can be verified with planetarium software such as Stellarium or by checking against Dennis Duke’s and Gerd Graßhoff’s lists of equatorial coordinates for the time of Hipparchus
   
   

In the conclusion:

1. it seems safe to assume nevertheless that, if his observations were conducted with an armillary sphere,
2. this must have been an equatorial armillary sphere, and not an ecliptic armillary sphere like Ptolemy’s;
3. however, it is also possible that the measurements were taken with a dioptra, which may have been easier to operate.

All this is wrong

The statement that there were boundaries for constellations in antiquity is certainly wrong. This can be clearly proven in the Almagest, and to make such a hypothesis for Hipparchus shows ignorance of his text. The Almagest uses the words ‘length’ and ‘width’ on the globe as we use ‘longitude’ and ‘latitude’: they are expressions for spherical coordinates and not for the boundaries of constellations, and the latter are explicitly not given: There is a paragraph describing the shape of the constellation surfaces. The translation from Greek is not entirely clear; it may refer to skeletal lines or outlines, but not to rectangular boundary lines by coordinates (as used today by the IAU).

We do not know the measurement uncertainty of Hipparchus. We only know that the scales of his globe allowed a reading accuracy of half a degree, i.e. his globe had a scale with 1° as the smallest part of the scale. We also know that he used data from different sources: in his part 3 he must have used a star list in addition to the globe, which had a 10x greater measurement accuracy and which consequently had a different source (Hoffmann 2017).

The proposed check of the data in the paper by Gysembergh et al. (2022) with Stellarium actually shows that his numbers do not match (new JHA paper).

About the Conclusions

Unfortunately, the rest of the assertion developed in the text by Gysembergh+ is not correct either:

• Sentence structure and ‘wording ’ (Greek vocabulary used) in the medieval manuscript do not match Hipparchus’s formulations
• the addition in the translation ‘to rise’ shows that the translators do not even compare the correct numbers, because Hipparchus does not give rises, but culminations in the numerical notation used here as proof.
• Of the four numbers handed down in the medieval text, only one can be reconciled with the epoch of Hipparchus – the other numbers point to other epochs (between -900 and +1400), so that the numbers in the manuscript are most probably wrong: it is reasonable to assume that the Greek text was deleted from under the Syriac text because it was incorrect and no longer needed.

So if not even the numbers in the text refer with certainty to Hipparchus and his era, then neither can they be used to indicate Hipparchus’ star catalogue, nor can anything be deduced about measurement methods and the transfer to Ptolemy.

Hipparchus’ star catalogue is said to have had roughly 850 entries. So to conclude from incomplete coordinates for three stars (four numbers!) that the catalogue has now been found again (which is claimed in the press) is boundless imposture.

The fact that Hipparchus was one of the sources for Ptolemy, but not the only one (which is presented as a new finding in the specialist article), has been scientifically proven for centuries (Tycho Brahe around 1600, H. Vogt in the 1920s, Grasshoff in the 1980s) and is even stated expressis verbis in the Almagest (137 AD). There is therefore no need for the new manuscript.

That Hipparchus observed with an armillary sphere, on the other hand, is not proven and is also not made clear by the new manuscript. On the contrary, I proved in 2017 that his data came from various sources – i.e. if, among other things, from an armillary sphere, then there are also other instruments/literature that were used.

Result

Thanks to the colleagues Gysembergh, his co-authors, referees and editors that even the cheap seats in the back row should now have realised that Ptolemy would not have fraudily stolen data from Hipparchus (what would be punished as scientific misconduct today and was not done back then either). He did what any good scientist would do: build on the results of others and cite them. However, this has been known for a long time and is not a new finding.

Thank you also for demonstrating how easy it is to pull the wool over the press’s eyes. For a paper whose interpretation is wrong from the beginning to the end, this one has received an alarming amount of attention (with even more falsehoods being spread in simplified press statements). My doctoral supervisor was quite angry – and rightly so.

Aftermath

Incidentally, the same thing happens on numerous other occasions: I could list several occasions where astrophysicists try to engage with ancient history or historical data, but I’ll limit myself to one example regarding Hipparchus:

Brad Schaefer’s claim in 2005 that the Farnese globe was supposedly based on Hipparchus’s star catalogue made headlines in public media, including Spektrum. However, the counter-paper by Dennis Duke (only six months later!) was not mentioned in the public and Duke completely dismantled Schaefer! This is how nonsense remains in the collective memory called the Internet and not the truth.

Sad!