Apollonius of Perga, Conics. Volume 1.1: Book I. Greek and Arabic text established, translated and annotated under the direction of Roshdi Rashed. Historical and mathematical commentary, edition and translation of the Arabic text by Rashed (Q644972): Difference between revisions

This is a major accomplishment of a group of historians of mathematics, presenting the reader with a magnificent edition of the Greek and Arabic original, as well as the French translation of the seven extant books of Apollonius's \textit{Conics}, a major accomplishment of ancient mathematics. The most important question one needs to answer before embarking on such a monumental task is to whom the work is addressed, what readership is one writing for? Is it for classical philologists, who now regard Greek mathematics as a ``literary genre'', or is it for the mathematicians of today and tomorrow? As \textit{B.~Artmann} wrote on p.~46--47 of [I.~Mueller (ed.), On mathematics. Essays on ancient mathematics and its later development based on a conference, Oberwolfach, Germany, August 19--25, 1990. Edmonton: Academic Printing and Publishing. Apeiron 24(4), 1--47 (1992; Zbl 0925.01006)]: ``Mathematicians tend to stress isomorphisms: they like to see the same structure in different guises; by contrast a philologist puts great value on expression and literary form. Mathematicians are accustomed to separating the content of a proposition from its form of expression, whereas philologists are likely to stress the particularity of different forms of expression.'' The editors of this \textit{opus magnum} have decided to address it to the mathematician, explaining the translated text for the modern reader in the language of algebraic geometry in the commentaries section of each book, fully aware of the ire they raise among the philologists, opening themselves up, in the directing editor's own words (p.~vii), to ``un reproche d'anachronisme de la part des gardiens du temple.'' The reproach did not fail to come, in the strongest terms this reviewer has \textit{ever} read, in the review \textit{F.~Acerbi} wrote of volume~2, with no trace of considerations of collegiality, in [Nuncius 26, 393--394 (2011)]. Why would a present-day mathematician want to read this work? What is there still to be said about conics in the 21st century, after all the work of Fermat, Descartes, La Hire, Desargues, Pascal, Poncelet, Steiner, von Staudt, after it has become a footnote in algebraic geometry or in the theory of quadratic forms? Since Apollonius develops the theory of conics in a synthetic manner, what one ought to do -- and of which there is still no trace to be found in the literature -- is to ask what the minimal axiomatic assumptions are under which Apollonius's many propositions hold. One may ask what remains valid of the theory if one starts with a cone in the 3-dimensional absolute geometry axiomatized by \textit{J.~Ahrens} [Math.\ Z. 71, 154--185 (1959; Zbl 0088.37003)] (with or without axioms of order, depending on whether the statement of the proposition requires the existence of an order notion). The statements of several propositions make sense only in the presence of a Euclidean metric, as they refer to parallels, but for those that can be stated in Ahrens's absolute setting, one can ask, in case the proposition does not hold in that general setting, what additional axiom (one that will likely fix the metric of the space) is that proposition equivalent to. A similar analysis could be made for the same propositions based on the plane characterization of each conic section, for which Apollonius provides a solid beginning in Propositions I.11, I.12, I.13, either inside the axiom system for metric planes [\textit{F. Bachmann}, Aufbau der Geometrie aus dem Spiegelungsbegriff. 2.~ergänzte Aufl. Berlin-Heidelberg-New York: Springer-Verlag (1973; Zbl 0254.50001)] (enriched, if necessary, with order axioms), or, in case parallels are a part of the statement, inside the axiom systems for Euclidean planes over skewfields (such as those proposed in \textit{H.~Naumann} [Math.\ Ann.\ 131, 17--27 (1956; Zbl 0070.15703)], \textit{H.~Naumann} and \textit{K.~Reidemeister} [Abh.\ Math.\ Semin.\ Univ.\ Hamb.\ 21, 1--12 (1957; Zbl 0077.13901)], or \textit{E.~Kusak} [Bull.\ Pol.\ Acad.\ Sci., Math.\ 35, 87--92 (1987; Zbl 0618.51001)]), and investigate whether their validity is equivalent to the validity of the Pappus axiom (i.e., if they hold only in Euclidean planes over commutative fields). For all such axiomatic investigations, the translation of the text not only in French, but also in a mathematical language understandable today, is a must. The current edition contains the Greek version of Eutocius of the first four books of the \textit{Conics}, as well as the text of an Arabic translation of the first seven books of the \textit{Conics}. The story of manuscripts, multiple Arabic translations, is told in great detail, and Rashdi Rashed proposes a criterion of mathematical clarity and of superior presentation not only for selecting the Arabic translation, but also to put forward a hypothesis of several editions of the \textit{Conics}, which had circulated during Apollonius's times, a preliminary one and a revised one, and of the hypothesis that these differences can be read off the different versions of the Arabic translations. Given that no less than Thābit ibn Qurra was involved in the translation of one of the versions, establishing the original (and lost) version of the \textit{Conics} by the criterion of mathematical clarity will not be accepted by historians of Greek mathematics, and establishing the (closer to the) \textit{original} does not seem to be the kind of issue that will ever be resolved (an explanation for the reasons why such an undertaking has little chance of success can be found in \textit{N.~Sidoli} [Isis 102, 537--542 (2011; \url{doi:10.1086/661629})], an essay review of this translation). Regardless of one's views on the matter of the success in having recovered the better and historically more accurate version of the \textit{Conics}, this is truly a major achievement.