The work contained in this volume of Ancient Commentators on Aristotle is not ancient at all: it is an extract from Michael of Ephesus’ twelfth-century replacement for the lost part (on books Ε–Ν) of Alexander of Aphrodisias’ commentary on Aristotle's Metaphysics. The supplement was transmitted together with Alexander on Α–Δ, and with time everything came to be attributed to Alexander, whose authorship was unchallenged till the nineteenth century. M. Hayduck's edition in Commentaria in Aristotelem Graeca is entitled Alexandri Aphrodisiensis in Aristotelis Metaphysica Commentaria (1891), and one has to read the Latin preface to realise that Ε–Ν are not by Alexander. The attribution to Michael has a strong basis, and M., who in his introduction presents a fine survey of the debate about the authorship, seems to believe in it, but to be on the safe side he calls the author ‘Ps.-Alexander’. I shall say ‘Michael’.

M.'s introduction also outlines the challenges that face any commentator on book Λ and offers an overview of how Michael met them. M. is aware that Michael pillaged earlier sources, and probably more than we can identify, but he makes a brave attempt to convince the reader that the result is decently coherent. Finally, he presents what is known about the manuscript tradition of the commentary and about the relation of Michael's lemmata to the two branches of the Aristotelian text. He mentions that Hayduck failed to use the famous Oceanus MS, but does not appear to have consulted it himself.

In general, M.'s translation is clear, and the English flows naturally. Moreover, he has enlarged Michael's lemmata, so that the reader can see what is being commented on. Copious notes provide further help, and there are useful indexes and appendices, including one that translates J. Freudenthal's 1884 demonstration that Averroes’ quotations of Alexander on Metaphysics Λ prove that what M. has translated cannot possibly be the Aphrodisian's work.

An examination of selected passages has convinced me that, generally speaking, the translation is not only readable but also reliable. There are, however, some slips in the treatment of Greek idioms, niceties of syntax and, in one case, a logical rule.

1. 1) At 666.8–12 (Hayduck's page and line numbers) M. translates:

   Alternatively, since he had enquired in the [book concerning] puzzles (aporia) in addition to other issues as to whether all things have the same principles or different things have different ones, but has not yet hit upon a solution (lusis), he for this reason again takes up the account of causes (aitia) in order to prove (deiknunai) that in one way all things have the same principles but in another way they do not.

   The Greek is:

   ἄλλως τε ἐπειδὴ ἐν ταῖς ἀπορίαις πρὸς τοῖς ἄλλοις καὶ τὸ πότερον πάντων αἱ αὐταί εἰσιν ἀρχαὶ ἢ ἄλλων ἄλλαι ἐζητεῖτο, οὐ τετύχηκε δέ πω λύσεως, διὰ τοῦτο πάλιν τὸν περὶ τῶν αἰτίων προχειρίζεται λόγον, ἵνα δείξῃ ὅτι τρόπον μέν τινά εἰσι πάντων αἱ αὐταὶ ἀρχαί, τρόπον δέ τινα οὔκ εἰσιν.

   M. mistranslates ἄλλως τε ‘furthermore’ as ‘alternatively’ and misconstrues τετύχηκε. I think ‘prove’ for δείξῃ is too strong, but that is a minor matter. I would translate the passage as follows:

   Furthermore, since in the puzzles one of the several questions raised was whether all things have the same principles or different things have different ones, and that [question] has not yet received a solution, he again takes up the discussion about causes in order to show that in one way all things have the same principles while in another way they do not.

2. 2) At 685.27–9 Michael wrote:

   Ἐντεῦθεν περὶ τοῦ πρώτου αἰτίου καὶ τῆς πρώτης οὐσίας, ὃν καὶ θεὸν ἐνταῦθα καλεῖ, διαλέγεται. συνίστησι δὲ πρῶτον ὅτι ἔστι τις τοιαύτη οὐσία ἀίδιος οὕτως.

   This means:

   From here on he discusses the first cause and first substance, which he here also calls ‘god’. First he establishes that there is such an everlasting substance, as follows:

   M. writes:

   From here on Aristotle discusses the primary cause and the first substance, which in this work he also calls a god (theos). First of all he establishes that such a substance is everlasting in the following way.

   I find the indefinite article in ‘a god’ misleading, but this is debatable. However, rendering ἔστι as a copula is definitely wrong, as also shown by the sequel (which M. translates correctly).

3. 3) 715.9–13:

   ὥσπερ γάρ, φησί, τὸ στράτευμα ἔχει καὶ ἐν τῇ τάξει τὸ εὖ καὶ ἔστιν αὐτῷ ἀγαθὸν καὶ ἄριστον ἡ τοιάδε τάξις, ἔστι δὲ τῷ στρατεύματι εὖ καὶ ἀγαθὸν ‘καὶ ὁ στρατηγός, καὶ μᾶλλον οὗτός’ ἐστι τὸ εὖ (διὰ γὰρ τὸν στρατηγόν ἐστι καὶ ἡ τοῦ στρατεύματος τάξις …

   My translation:

   just as, he says, the excellence of an army consists partly in its order, and a certain order is what is good and optimal for it, ‘but the general is also’ what is the excellence and good of an army, ‘and he is even more’ its excellence, since the order of the army is due to the general …

   Failing to recognise that the first καὶ prepares for ἔστι δὲ … καὶ, M. produces four co-ordinate clauses: ‘an army possesses … and …, and …, and …’, as follows:

   just as, he says, an army possesses excellence in its order and a certain sort of order is good and best for it, and ‘the general’ (stratēgos) is excellent and good for the army (strateuma), and the general is excellent ‘more so’ than the army, since the order of the army is due to the general …

   The choice of ‘excellence’ for τὸ εὖ is a happy one, but M.'s translation does not catch the import of the neuter gender of ἀγαθὸν and ἄριστον.

4. 4) In Byzantine commentaries, ἤτοι is standardly used in the sense of ‘i.e.’ to introduce a gloss on a preceding word or phrase. In footnote 85 M. recognises this, but claims that ‘literally’ ἤτοι means ‘or’. Mostly, he correctly translates ‘that is’, but occasionally he uses ‘or’, and this happens whenever Michael glosses εἶδος with ἰδέα. This variation is unjustified. All of Michael's 70-odd ἤτοι are epexegetical.

5. 5) M. has troubles with εἰ ἔστιν ἡ φιλία … τὸ ἀγαθόν, τὸ νεῖκος ἔσται κακόν· εἰ γὰρ τὸ ἐναντίον τῷ ἐναντίῳ ὑπάρχει, καὶ τῷ ἐναντίῳ τὸ ἐναντίον at 718.15–18. In footnote 494 he claims that literally εἰ γὰρ etc. is a tautology, but that the following is ‘evidently the meaning’:

   if Love is … the good, then Hate will be bad, for if the former [i.e. good and bad] are contraries, the latter [i.e. Love and Hate] will be too

   This is not quite so. Michael is quoting a topical rule, first evidenced in Alexander of Aphrodisias, Comm. Top. 126.16–17, which means ‘if A and B are contraries, and C and D are contraries: if one item from the second set is true of one from the first set, the remaining member of the second set is true of the remaining member of the first set’, i.e. C(A) → D(B). D(A) → C(B). C(B) → D(A). D(B) → C(A). In casu, Love and Hate instantiate A and B, good and bad C and D; hence ‘Love is good’ – C(A) – implies ‘Hate is bad’ – D(B).

There are other minor slips. Thus, ἀνδράποδον has become andrapodos, and several Greek verbs that occur in the passive voice are indexed under their active form, resulting in entries like ‘apallattein, be devoid of’.

This being said, I can still recommend the translation, for, on the whole, I believe, it can be trusted to convey the meaning intended by Michael.