User:Syncategoremata/Elliptical orbits

There are various claims made on Wikipedia about the discovery that planets move in elliptical orbits (whether geocentric or heliocentric ones).

This is claimed mostly for
 * Āryabhaṭa
 * al-Zarqālī
 * Bīrūnī

I have no particular expertise in the case of Āryabhaṭa, but I will discuss the other two cases here.

For more details on much of this area, see also Talk:History_of_astronomy/Common_misconceptions put together by. I also use some material from his sandbox.

Abū Ishāq Ibrāhīm al-Zarqālī
See Talk:History_of_astronomy/Common_misconceptions for many sources discussing al-Zarqālī's writings on the orbit of Mercury, which is the only one for which there is any serious claim about it being elliptical. The paper by Julio Samsó and Honorino Mielgo (1994) discussed on that page shows pretty conclusively that although al-Zarqālī treated Mercury's deferent as an ellipse but only so far as this allowed an easier way to draw the figure concerned for an equatorium. The mechanisms of the orbit were entirely Ptolemaic and none of his calculations were ever based on an ellipse, but always followed Ptolemy's system from his Handy Tables.

Against these recent scholarly sources, most claims about al-Zarqālī and elliptical orbits come from much older or from non-scholarly sources.

References used in Wikipedia in support of the claim for elliptical orbits
On page 190, Briffault says that "Although the complexity of the Ptolemaic system was repeatedly criticized by Moorish astronomers, although Al-Zarkyal declared the planetary orbits to be ellipses and not circles, although the orbit of Mercury is in Al-Farȃni's tables actually represented as elliptical, although Muhȃmmad Ibn Mȗsa glimpsed in his works on Astral Motion and The Force of Attraction the law of universal gravitation, those adumbrations of the truth were not fruitful of any great reform."
 * Briffault 1919

There are problems with the reliability of this source though they are usually a fault of Briffault's sources, which he reports fairly well (see my User:Syncategoremata/Ibn Sīnā and the invention of the thermometer for some discussion).

This is a brief historical summary, written for a popular magazine and gives no sources. It is a pretty fair summary of the relevant material, but the following sentence is entirely back-to-front: "In this work [sc. the Alfonsine tables] Mercury's orbit is represented by an ellipse; geocentric of course, but interesting as the first representation of the motion of a heavenly body that departed from the Greek idea of uniform circular motions. (pp. 236–237)"
 * Rufus 1939

Not only did Ptolemy's theory already "depart from the Greek idea of uniform circular motions" (for which he was repeatedly chastised by Islamic astronomers), the sources given at Talk:History_of_astronomy/Common_misconceptions clearly show that al-Zarqālī is just following Ptolemy's theory.

The 'historical' introduction to this book claims that al-Zarqālī thought all the planets moved in ellipses (pp. 5–6). This is an extremely poor source and should not be used for any such claims. As an example of its quality, it claims that, "The Arabs discovered [the planet] Mercury" (p. 6).
 * Qadir 1989

Abū Rayḥān Bīrūnī
The consensus view on Bīrūnī and elliptical orbits for the planets is plainly that he believed no such thing. The strongest claims that can be made for him in this area seem to be the following:
 * He held that it was conceivable that the heavens themselves were elliptical though their rotation would still be circular, i.e., their overall shape could be a spheroid. This comes from the sixth question of his debate with ibn Sīnā (and has been partially misconstrued in Nasr (1993); see below).
 * He held that it was conceivable that the planets did not have an inherent principle of uniform circular motion but moved in forced orbits (this is from question one of the same debate). I can find no reference to his claiming that the motions of the planets is elliptical, for instance, and this issue appears to be one about Aristotelian physics rather than astronomy.

Question six
The usual support for the claim that Bīrūnī proposed elliptical orbits for the planets is p. 19 of: which is from question 6, "On the possible elliptical shape of the heavens" from his al-As¹ila wal-Ajwiba ("questions and answers", a work recording an exchange between Bīrūnī and ibn Sīnā) quoted from the partial translation on pp. 133–138 of:

The full text of the question is: "[Aristotle] has mentioned in Book II that [the shape of the heaven is of necessity spherical because] the oval and the lenticular shapes would require space and void whereas the sphere does not, but the matter is not so. In fact, the oval [shape] is generated by the rotation of ellipse around its major axis and the lenticular by its rotation around its minor axis. As there is no difference concerning the rotation around the axes by which they are generated, therefore none of what Aristotle mentions would occur and only the essential attributes of the spheres would follow necessarily. If the axis of rotation of the oval is its major axis and if the axis of rotation of the lenticular is its minor axis, they would revolve like the sphere, without needing an empty space (makan khal). This could happen, however, if the axis of [rotation of] the oval is its minor axis and the axis of [rotation of] the lenticular is its major axis. In spite of this, it is possible that the oval can rotate around its minor axis and the lenticular around its major axis, both moving consecutively without needing an empty space, like the movement of bodies inside the celestial sphere, according to the opinion of most people. And I am not saying this with the belief that the celestial sphere is not spherical, but oval or lenticular; I have tried hard to refute this theory but I am amazed at the reasons offered by the man of logic."

But this question is purely about the shape of the entire heavens and that the argument Aristotle gives for the heavens being spherical (so that it can rotate in place without needing a space within which to rotate) works just as well for an ellipsoid of rotation (spheroid) rotating about the relevant axis. Note that this clearly implies that in the plane of rotation the heavens are still circular, just as for the standard Aristotelian account, since only a circular rotation avoids the need for an external space within which the rotation can occur. And as Bīrūnī himself says in the question, he is not even claiming that the heavens are not spherical.

Covington 2007
There is also a popular science article in an online magazine published by the Saudi Aramco oil company:

This article gives no sources and is written by a science journalist. See Talk:Astronomy in medieval Islam for some good reasons not to use this article as a source here.

Qadir 1989


The 'historical' introduction to this work is sometimes used to support this but in fact it contains no such claim. It does contain a claim about al-Zarqālī for which, see above.

Sources checked but which do not support the claim
The following discussions of Bīrūnī's astronomy fail to mention anything about elliptical orbits. See also Talk:History of astronomy/Common misconceptions for more material about the Islamic astronomers' general adherence to Ptolemaic principles and for three sources not discussed here: Hartner 1955, and Kennedy 1970 and 1971, all of which show Bīrūnī to have used circular motions in this astronomical work.


 * Nasr 1993

This is cited in support of the claim that "Several Muslim astronomers also discussed the possibility of a heliocentric model with elliptical orbits". A more accurate report of what is in this edition at least should include no mention of elliptical orbits and that just Bīrūnī considered the possibility, based on either Greek or Indian sources, though he eventually settled on a geocentric model due to physical considerations. (See p. 136–136). The book is clear that "al-Bīrūnī adopts the Ptolemaic system of epicycles with later accretions and refinements added by Muslim astronomers" (p. 136).

Before quoting from question six (as above), Nasr does say that Bīrūnī implies that "the heavens could have an elliptical motion" (p. 170), which is clearly not what the text says. It also plainly contradicts what he wrote on page 136, as I have quoted here. See the following review of another of Nasr's works for some strong criticism of his handling of such claims:


 * Dallal 2001

Discussing the work al-As¹ila wal-Ajwiba (questions and answers) mentioned above, Dallal says: "In the course of this exchange, Biruni questions almost all of the fundamental Aristotelian physical axioms: he rejects the notion that heavenly bodies have an inherent nature, and asserts that their motion could very well be compulsory; he maintains that there is no observable evidence that rules out the possibility of vacuum; he further asserts that, although observation corroborates Aristotle's claim that the motion of heavenly bodies is circular, there is no inherent "natural" reason why this motion cannot be, among other things, elliptical."

So, another claim that Bīrūnī held the orbits to be circular.


 * Yano 2007

No mention of planetary orbits.

The orbit of Mercury
s.v. "ʿUṭārid", Julio Samsó, EI2, vol. 10, pp. 940b–942b.