The volume under review, Philoponus’ Commentary on Aristotle’s Posterior Analytics 1.9-18, is McKirahan’s second contribution to the translation of the whole of Philoponus’ commentary in the Ancient Commentators on Aristotle series. With this volume published, we can now look forward to the forthcoming volume by Goldin and Martijn, with which the entire work will be completed.1 The translators’ efforts, and McKirahan’s especially, will no doubt be greatly appreciated in the academic community and will feed the continued and ever-increasing interest the Posterior Analytics, as well as late ancient commentators, have recently enjoyed.
Accompanied by an introduction, over 500 footnotes, a bibliography and glossary, as well as indices covering words, passages, subjects and persons, this translation will surely serve as a good starting-point for engagement with Philoponus’ thought and will open up this interesting, if controversial, take on Aristotle’s Posterior Analytics for those who have little or no Greek. Some points, however, left me unsatisfied.
The short introduction, which, as the author himself admits, has in large parts been taken from the first volume, contains (apart from a section on certain conventions adopted by him, on which more later) some general background information concerning the commentary, possible sources of influence, some notes on its reception and a large section dedicated to pointing out noteworthy features of the text. This last part is certainly the most interesting and helpful part of the introduction, and draws the reader’s attention to aspects that might well be worth further investigation. Why, for instance, is there a significant change of style from the eleventh chapter onwards? And how does Philoponus conceive of the relation between the theory of demonstration and first philosophy? A feature McKirahan is particularly interested in is “Philoponus’ idiosyncratic treatment” of Aristotle’s views on scientific principles (4 ff.).2 In Posterior Analytics I.2 (72a14-21), Aristotle introduces three kinds of scientific principle: axioms, definitions and hypotheses. In his interpretation, Philoponus chooses to take a later passage (I.10, 76b23-34) into account, in which Aristotle distinguishes hypotheses from postulates, and identifies the concept of hypothesis mentioned there with that mentioned in the earlier passage. This, according to McKirahan, is a grave mistake and has “catastrophic consequences”; for, since indemonstrability is a key property of scientific principles, and since in the later passage hypotheses are characterised as demonstrable, an identification of the two should be ruled out immediately. It is interesting to note that Proclus in his Euclid commentary, a work with which Philoponus was certainly well-acquainted, seems to be guilty of a very similar ‘contradiction’, for he characterises postulates as indemonstrable scientific principles while at the same time calling one of them (the fourth) demonstrable. However, at least in the case of Proclus, an argument can be made to show that this is not a contradiction and has no catastrophic consequences at all.3 For if we allow ἀναπόδεικτον to mean ‘not requiring demonstration’ instead of ‘indemonstrable’ or ‘undemonstrated’, we can understand scientific principles as intrinsically not requiring demonstration. While some might really be indemonstrable, others are demonstrable and are indeed demonstrated when the need arises – for instance, when a teacher has to deal with a student who has problems comprehending Euclid’s fourth postulate. Whether or not a defense on these lines can be made for Philoponus remains to be seen, but there is some evidence that this is what he had in mind: Philoponus, just like Proclus, puts a considerable emphasis on the dialectics of teacher and learner and the different attitudes a learner can take towards propositions – including principles – put before him. In particular, Philoponus tries to distinguish principles insofar as we know them “from within ourselves” (οἴκοθεν) or not,4 and insofar as they are common by nature to all men or not.5 It is easy to see how these features chime in with Proclus’ distinctions of ἀναπόδεικτον.
Concerning the translation itself, I did not run into any obvious mistakes or clear misrepresentations. I have, however, a general complaint concerning McKirahan’s practice of translating forms of the words ἐπίστασθαι, γιγνώσκειν and εἰδέναι. McKirahan informs us in his introduction (8) that: “since [these three words] can all be translated as ‘know’, I distinguish them as follows: epistasthai is ‘know (e) ’ or ‘have scientific knowledge (e) ’, gignôskein is ‘know (g) ’, and eidenai is ‘know o ’.”6 It is of course true that all these words can mean ‘to know’, but this fact alone certainly is not a good enough reason to translate them all in this way. Indeed, it seems that there are two possibilities: either Philoponus does not distinguish between ἐπίστασθαι, γιγνώσκειν and εἰδέναι, regarding these words as synonyms: in this case the indices are superfluous and a simple translation with ‘know’ is sufficient to render the meaning. Alternatively, they may not be synonymous and Philoponus does distinguish between them; in this case, however, the indices are not sufficient for translating the text. Rather, the translator should have chosen an English word or phrase to express the meaning Philoponus intended. Furthermore, McKirahan makes no attempt to explain whether or not, and if so how, the meaning of the three words might vary. As a result, the indices cannot help a Greekless reader, and those who have access to the Greek text are left with the task of coming up with an appropriate translation themselves.
Another, though less important, issue is the inconsistency between McKirahan’s two volumes and those translated by the other contributors: Goldin (2009) adopts slightly different conventions (cf. p. x) and Goldin and Martijn have decided against the usage of indices altogether.7
I will now turn to what in my opinion is the most unfortunate characteristic of this volume, namely McKirahan’s eagerness to show that Philoponus’ commentary is – at best – a third-rate work. McKirahan makes no secret of this right from the beginning, when he tells us in the introduction that the commentary is not an important philosophical work in its own right (1), that Philoponus had little or no talent for mathematics and logic (3), that he uses Euclid and Proclus uncritically to explain Aristotle (6), etc. To be sure, it is the duty of a modern commentator to point out such shortcomings, but reading McKirahan’s notes I cannot help but feel that Philoponus is at times treated unfairly and that at other times employing the principle of charity would have been called for.
Some examples (more of which could be added):
1. At 128,5 ff. (p. 31), Philoponus paraphrases Euclid’s definition of the circle and then adds that this definition will be clear to “anyone who only understands that the distance from the centre, where one end of the compass is placed, to the other end is a straight line which generates the circle by being turned completely around.” In the corresponding footnote (115) McKirahan remarks that “Euclid’s definition of circle makes no reference to the use of a compass”. Yet neither does Philoponus’. As is evident from the text, Philoponus is talking about the relation between the different kinds of principles and the learner who tries to grasp them. Just a few lines earlier Philoponus explains that, while some principles, to which definitions belong, “require some attention, still they are true and it immediately becomes clear that they are so to a person who thinks about them for even a little.” The function of the compass, then, is clearly an example of this “thinking a little” and Philoponus cannot be blamed at all for conflating the definition of circle with the use of the compass, as McKirahan’s note implies.
2. Another example involves the translation itself. 120,19 ff. (at p. 24) runs thus: “The principles in each science (e), he says, are the things (αὗται) that (ἅς) it is not possible to demonstrate, but are granted. For example point, line, and so forth are principles of geometry.” In the corresponding footnote (54), McKirahan criticises Philoponus for conflating ‘things’ with definitions of and assumptions about them. However, the word ‘things’ is McKirahan’s addition: the text merely has ‘αὗται’, the demonstrative pronoun that corresponds with the following relative pronoun ἅς. McKirahan feels entitled to translate ‘things’ (taking ‘things’ not in its general, but in the specific sense criticised in the note) because of Philoponus’ examples. As an alternative, he could have assumed that Philoponus merely writes – as is surely not uncommon among ancient authors – in a brachylogical style, and that he actually meant not, e.g., the point itself, as a thing, but the definition of a point, or the assumption that it exists. Indeed, the sentence immediately following corroborates this reading, for here Philoponus continues his examples of what a specific science, in this case geometry, does not demonstrate: “It is not geometry’s task to prove that points are without parts . . .” Not only is it now clear that Philoponus is not thinking of terms but of propositions, but it is also evident that he is thinking of definitions, for we are clearly presented with a paraphrase of Euclid’s definition of the point.
Finally, I feel obliged to make a remark about the quality of the material book (I have checked several copies). In a time in which the book industry is under constant and ever-increasing pressure by the rise of alternative forms of publishing, the very least one should expect from a printed book, and even more so if it costs $90, is impeccable production quality. The book is nevertheless full of production errors. In fact, only few pages can be found that are free of errors: many letters look as if they had been cut into two unaligned parts (e.g. 33), the result being that they appear shadowed and as if set in bold type. Moreover, the geometrical diagrams are strongly reminiscent of early 9- pin dot-matrix printers: the resolution is so low that individual pixel-blocks can be seen with the result that ellipses are jagged (e.g. 17) and at times so too are straight lines (20). Of course, these problems don’t make the book unreadable, but if one compares the volume of Wallis’ text, 100 years older, one cannot help but recognise the decline of quality. If the reason for this decline is related to the costs of production, we might wonder why new forms of publishing are not taken up on a larger scale in academia.
Apart from the above-mentioned issues, this book is an important publication that will certainly help to make the work of Philoponus better known. The translation – the first ever into any modern language – makes the Greek text accessible to those with little or no Greek and will be an indispensable tool for students and scholars of ancient thought. The notes clarify many strange passages and provide valuable information about the tradition Philoponus is working in. Moreover, the book is well edited.8 Thus, there can be no doubt that many scholars and students will find the book highly useful.
1. The commentary on Book 2 was translated by Goldin and published in 2009.
2. McKirahan deals with this subject at length in McKirahan, ‘Philoponus’ Account of Scientific Principles in his Commentary on Aristotle’s Posterior Analytics’, Documenti e Studi sulla tradizione filosofica medievale 20 (2009), 211-63.
3. I have argued this in a paper given at the Proclus Workshop in Amsterdam, 02.02.2012.
4. McKirahan, in his quotation (6) writes autothen (as in his 2008 introduction, but there with the typo authothen). Philoponus in both passages (34,10 ff. and 127,21 ff.) has oikothen, and neither in Wallis’ apparatus nor in McKirahan’s emendations can autothen be found. McKirahan (2009: 239) has oikothen as well.
5. 127.24 ff. This, admittedly, seems quite strange: how can a certain principle be known to some people by nature, but to others (who supposedly have the same nature) not? If we do not want to attribute such a strange theory to Philoponus we might try to understand συμπεφυκυῖαι not as meaning that the knowledge of the principles in question is innate and a part of the individual’s nature. συμπεφυκυῖαι can also mean “having become assimilated, having become natural” (cf. LSJ, ad loc.), so that Philoponus’ choice of wording can be understood as motivated by his aim to show the above-mentioned distinctions of ἀναπόδεικτον: the self-evidence (αὐτόπιστον, 127,22) of a particular principle might be immediately obvious to one person (and then he can be said to have assimilated it, or that it has become natural to him), but not to someone else. This reading seems to chime in well with the context of the passage.
6. A special case, nowhere explained, can be found at 25, where McKirahan renders a form of προγιγνώσκειν with “to know previously (pg) ”.
7. I have consulted M. Martijn concerning the conventions of the forthcoming volume.
8. I could only spot one typo: 141,1 instead of 114,1 in the margin of p. 18.