Golitsis’ new edition of Alexander of Aphrodisias’ Commentary on Aristotle’s Metaphysics in the series Commentaria in Aristotelem Graeca et Byzantina is an excellent work that will certainly replace the hitherto standard edition of Hayduck from the original Berlin edition of Commentaria in Aristotelem Graeca in the late 19th century. This is not primarily due to the comprehensive and useful introduction (especially compared to Hayduck’s CAG edition), but to the corrections and additions made to the text based on a study of a wider set of manuscripts and a corrected reconstruction of the textual tradition.
The need for a new edition stemmed from the fact that ‘the previous editions are based on inadequate recensio et examinatio codicum’ (CXLIV). Golitsis makes his significant corrections on the textual tradition based on an examination of a significantly larger scope than previous editors, examination of almost all (20 of 23) the available manuscripts.
Most importantly, he reconstructs the stemma codicum anew. First of all, he gives the hitherto neglected manuscript O (Laurentianus pluteus. 85.1) its due place in it (XLVII–XLVIII). Being an independent direct witness to the text, O is used along manuscript A (Parisinus Graecus 1876) in reconstructing the text for the first time. Along with O, he reserves an important role to P (Parisinus Graecus 1878), insofar as it contains a number of authentic readings and useful variations (Section 2.1.1). This is the case despite the fact that P is not independent of O and A, rather, ‘P is an indirect copy successively of O (for the first three quires) and A (for the rest)’ (LVIII). The complex status of the manuscript is due to the scribal and editorial work of two learned scholars of the 15th century.
More than that, he convinces us that the alternative reading supposed by Hayduck (recensio altera) found chiefly in manuscript L (Laurentianus pluteus 87,12) is instead to be used as an indirect witness to Alexander’s commentary (XLVIII–LIV, cf. CXLIX–CLI). For, as he shows through its relation to Michael of Ephesus’ commentary on books Z–N of Metaphysics, the so-called recensio altera is rather an independent Anonymous commentary (using Asclepius, besides Alexander, as a source). Still, some parts of L seem to have been amended by Alexander’s text, which makes these parts to be a direct witness to Alexander’s commentary.
At the same time, Golitsis reduces the importance of Sepúlveda’s 1527 Latin translation for the edition. The translation is based on four allegedly old manuscripts, which, however, Golitsis demonstrates to be produced shortly before Sepúlveda’s time (CXLVI–CXLIX). These manuscripts are shown to depend mainly on A, and for some corrections on O, P, and the Anonymous commentary (the so-called recensio altera). Thus, the Latin translation is related to all the important branches of the textual tradition in one way or another, which explains the importance of the translation for previous editors working with an incorrect view of the tradition. Yet, despite the reduced importance of this translation for the edition, Golitsis praises Sepúlveda and accepts a number of his conjectures.
The text is reliable, as it might be expected from the above description. Apart from the corrections and additions (based on fresh collation or the use of additional manuscripts) Golitsis offers quite a few conjectures, including some significant ones. Page 4 provides representative examples for such editorial changes. Golitsis corrects previous editors’ μαθητικώτερα to μαθηματικώτερα throughout the five occurrences (as lectio difficilior, as well as based on agreement between A and P with Asclepius’ testimony (216, note 7)). Again, he adds ὡς to line 4.13 (preferring manuscript O and Sepúlveda over manuscript A), which results in a text with a better meaning, and closer reproducing Aristotle’s text (Mem. 1.451a14–16):
ἔστι δὲ μνήμη ἕξις φαντάσματος ὡς εἰκόνος οὗ ἐστι φαντασία· οὐ γὰρ ἱκανὸς πρὸς μνήμην ὁ τύπος ὁ κατὰ τὴν φαντασίαν, ἀλλὰ δεῖ τὴν περὶ αὐτὸν ἐνέργειαν καὶ ὡς περὶ εἰκόνα γίγνεσθαι, τουτέστιν ὡς ἀπ’ ἄλλου γεγονότος (4.11–13 Golitsis = 3.15–18 Hayduck)
Memory is having a phantasma which is like an image of that about which the phantasia is. The impression according to the phantasia is not sufficient for memory, but the activity concerning the impression must also be concerned as with an image, that is, it must be as from something else that has happened. (Dooley’s translation modified)
Finally, a previous lacuna is filled with an ingenious conjecture in 4.19–20 Golitsis (= Hayduck 4.3–4), providing a dog and a swallow as examples for animals capable of distinguishing what is their own beyond remembering: ὡς ὁ κύων καὶ χελιδών· ταῦτα […]
The text is also easily readable thanks to the well-structured print layout. Golitsis prints the lemmata as found in the manuscripts (more thoroughly than previous editors), and helps orientation in the commentary by indicating the commented passages of Aristotle on the left margin. Navigation is assisted also by the book and chapter numbers in the header. Due to the Hayduck numbering added to the right margin (Golitsis’ on the left) the edition is usable with old references to Hayduck’s edition as well.
Among the three apparatuses, the critical apparatus and the concordance between Alexander and Aristotle’s text are quite accurate and complete regarding the relevant variants. The apparatus fontium et referentiarum mainly records quoted passages and a few references to Aristotle, especially to the Metaphysics. For the benefit of those interested in interpretation and history, it could have been supplemented with more indirect sources, several of which are established in translations of Alexander’s commentary, like Dooley’s. Indeed, Dooley’s references can even supplement identifying explicit references missed in Golitsis’ apparatus, e.g. at 29.9 Golitsis (= 35.3 Hayduck) the reference of ὃ καὶ αὐτὸς ἐξηγήσατο is plausibly Poetics 1454a37–b2.
At the end of the volume there are indices of names and passages for the Introduction and the end-notes; indices for Alexander’s text will appear in the second volume. A few codex photocopies appear as illustrations of scribal and manuscript editorial work. After the text we get five appendices: (A) and (B) report interpolations to manuscripts A and P respectively; (C) lists peculiar readings of P (in addition to what is recorded in the critical apparatus and Appendix B); (D) contains readings of manuscript P that are adopted in the text; and (E) lists Golitsis’ original corrections and conjectures. Among the appendices, D and E give a sense of the difference between Golitsis’ text and the previous one of Hayduck’s. This, however, leaves us with a partial view and does not provide an easy way of determining the differences. It would have been quite useful to get an index of differences between this edition and Hayduck’s—the only thing missing from the book—which presumably will be published in volume two.
The text is preceded by an introduction of 150 pages (in English) covering all aspects of the work, the bulk of which relates to the textual tradition (Chapter 2–4) and the relationship of the edition to previous editions (Chapter 5). Yet, the very first Chapter 1 concerns interpretation. Golitsis advocates attributing a unitary reading of Aristotle’s Metaphysics to Alexander (Section 1.1). Accordingly, the science of metaphysics, first or universal philosophy, concerns both primary substances—namely, immaterial forms or intellects (most of all God, the active intellect) that are the first movers (as final causes) of secondary substances, the celestial bodies in perpetual circular motion, that are, in turn, principles of substances in the sublunary world—and concerns being qua being, that is, being without any determination, and so covering any kind of formal aspect of being. As this science is chiefly about God as primary intellect and primary substance, it is theological, hence first philosophy culminates in Metaphysics Lambda.
Golitsis also helpfully summarises Alexander’s commenting method (Section 1.2). Relying on an established commentary tradition on Metaphysics (Aspasius’ commentary and anonymous scholia) Alexander reports and criticises—apart from the content—textual variations, which is quite helpful in establishing Aristotle’s text. Alexander’s commentary stands, however, relatively early in the tradition, hence it does not follow strict formal rules. One such rule is starting with some general considerations about the whole work that is being commented, namely with a prolegomena. Alexander’s prolegomena is apparently missing from the manuscripts (which start abruptly, due likely to a lost folio of the archetype), yet Golitsis is able to provide one, based on Asclepius’ commentary, which is largely indebted to Alexander’s, and the considerations of which reappear in other commentaries of Alexander and the prolegomena to them. With this, Golitsis extends Hayduck’s edition with 36 lines.
In sum, Golitsis provides an edition superseding previous editions with a reliable text of Alexander’s commentary based on critical study of the available sources. Having the text, perhaps it is time to revise the translations of Dooley and Madigan in the Ancient Commentators on Aristotle series, especially as they appeared between 1989–1993 among the first few volumes. Nonetheless, from now on any reference to Alexander’s commentary should be made to Golitsis’ edition using his pagination.
 Hayduck, M. 1891. (Ed.) Alexandri Aphrodisiensis in Aristotelis metaphysica commentaria, consilio et auctoritate Academiae Litterarum Regiae Borussicae. Berlin.
 The importance of O is already emphasised by Dieter Harlfinger, “Edizione critica del testo del ‘De Ideis’ di Aristotele”. In: W. Leszl (ed.) Il ‘De ideis’ di Aristotele e la teoria platonica delle idee. Florence. 1975, p. 22–39.
 See also P. Golitsis, “Who were the real authors of the Metaphysics commentary ascribed to Alexander and ps.-Alexander?” In: R. Sorabji (ed.) Aristotle Re-Interpreted. London. 2016, p. 565–588.
 In most cases he helpfully explains the editorial decisions in textual notes.
 Dooley, W. E. 1989. Alexander of Aphrodisias on Aristotle’s Metaphysics 1. Cornell University Press, Ithaca.
 Notably, Dooley has already suggested dog as an example here (op.cit. 17, note 26), which Golitsis does not duly acknowledge.
 There are, among diverse kinds of divergences from Hayduck, for example, 79 corrections of mistaken readings (originating from Brandis’ transcription that Bonitz and Hayduck took over) marked in the critical apparatus as “perperam edd.” (CXLIV– CXLV).
 Including the detailed description of the extant manuscripts of the direct and indirect witnesses to the commentary (Chapter 3). One might think that since it will likely be infrequently used, the 45 pages it takes is not warranted, and despite its curiosity to a few, most of it could have been omitted from this edition and published elsewhere.
 In addition to Dooley’s op.cit., see Dooley, W.E. and A. Madigan. 1992. Alexander of Aphrodisias on Aristotle’s Metaphysics 2 & 3. Cornell University Press, Ithaca.
 I am grateful to Ágoston Guba for his suggestions for clarifying the text. The review has benefited from the financial support of the Hungarian National Research, Development and Innovation Office (NKFIH: project OTKA-138275).