The book presents a new critical edition and Italian translation of Book Lambda of Aristotle's Metaphysics, with an introduction and critical notes. The introduction comprises a first-rate reconstruction of the editorial history of the Metaphysics as well as a discussion of the editorial criteria followed in the critical edition, whereas the notes provide detailed justification for the editorial choices made and occasionally discuss difficult points of translation. The book is the first part of a larger project F. intends to complete in the near future, which includes, besides the new critical edition of the text, a philosophical commentary on Lambda. In this review, I will mainly focus on the critical edition and the editorial criteria, while leaving out of consideration the new (and generally accurate) Italian translation.

F.'s edition marks substantial progress with respect to the two prestigious twentieth-century editions of the Metaphysics by Ross (1924) and Jaeger (1957). To see why this is the case, it may be useful to summarise briefly what we know about the transmission of the text. Ross's and Jaeger's editions are based on three independent witnesses of Aristotle's text: Parisinus gr. 1853 (E: tenth century), Laurentianus plut. 87.12 (A^b: mostly twelfth century) and Vindobonensis phil. gr. 100 (J: ninth century), ‘re-discovered’ by Gerke in 1892. Gerke also showed that the tree manuscripts represent two independent families in the transmission of the text, with E and J being independent witnesses of the so-called α-family and A^b representing the β-family. Ross and Jaeger also make use of two indirect sources: the commentary by Pseudo-Alexander and the Latin translation of the Metaphysics by William of Moerbeke (late-thirteenth century). Since Ross and Jaeger published their influential editions, our knowledge of both the manuscript and the indirect tradition of the Metaphysics has been considerably enlarged. In 1979 Harlfinger (‘Zur Überlieferungsgeschichte der Metaphysik’ in P. Aubenque [ed.], Études sur la Métaphysique d'Aristote [1979], pp. 7–36) published a complete stemma codicum of the Metaphysics and made at least two fundamental discoveries: he identified two further independent manuscripts in the β-family, the recentiores Ambrosianus F 113 Sup. (M: fourteenth century) and Tauriniensis VII B 23 (C: fifteenth century); he proved that at Λ 7, 1073a1 (where there is a change of hand: a fourteenth-century scribe resumes where a twelfth-century copyist left off) A^b ceases to be a witness of the β-family and becomes a copy of a lost sub-archetype of the α-family. This means that, from Λ 7, 1073a1 onwards the recentiores M and C are the only independent witnesses of the β-text. Important discoveries have been made concerning the indirect tradition as well. As suggested by Praechter, Pseudo-Alexander's commentary on Books E–N has now been shown to be the work of the eleventh/twelfth-century Byzantine scholar Michael of Ephesus (C. Luna, Trois études sur la tradition des commentaires anciens à la Métaphysique d'Aristote [2001]). Moreover, Vuillemin-Diem (AL XXV 3) has demonstrated that William of Moerbeke's translation is based on J or a copy thereof, and so makes no independent contribution to the constitution of the text.

We are now in a position to see the innovative aspects of F.'s edition, which not only makes full use of the most recent discoveries on the manuscript and indirect traditions of the Metaphysics, but also makes an additional contribution of its own. There are at least four points that deserve to be mentioned here. (1) Although her edition is largely based on Harlfinger's stemma, F. calls into question one of the assumptions behind the stemma, i.e. that before the change of hand at Λ 7, 1073a1 A^b still represents the β-text. F. argues, by contrast, that the stemmatic position of A^b should be reconsidered, and more particularly that A^b ceases to be a witness of the β-family well before 1073a1 and more precisely at Κ 8, 1065a26. If she is right, then M and C remain the only witnesses of the β-text for the whole of Lambda and not only for the final section of the book. I think that F. is absolutely right that the stemmatic position of A^b in Lambda must be reconsidered and the evidence she offers (pp. 113–18) is overwhelming in this respect. I agree with Primavesi (Aristotle. Metaphysics Α [2012], pp. 395–6), however, that it is an exaggeration to claim that A^b ceases completely to belong to the β-family; the evidence F. presents rather points to A^b being contaminated by the α-family. F.'s slight overstatement has no significant consequences for the text that she publishes, as her edition rests on solid philological grounds as well as on a careful analysis of Aristotle's modus scribendi. In line with a suggestion by Frede and Patzig (Aristoteles ‘Metaphysik Zeta’ [1988], I, p. 16), she shows the undisputable superiority of E and J over A^b , which has a tendency to present an easier, plainer and normalised text, and follows therefore the variants of the α-family. Actually, F.'s analysis of the ‘editorial’ and normalising character of A^b 's text is particularly lucid and illuminating (pp. 118–28). Thus, whether or not the A^b variants should be discarded on a stemmatic basis only (as F. claims), they can certainly be cast aside (in virtually all cases) on the basis of general philological considerations; and F. is certainly right to say that the authority which Ross and especially Jaeger credited A^b with is undeserved. (2) F. is the first to make use of manuscripts M and C. This is an objective merit of the edition, which is particularly evident in the portion of the text (after 1073a1) where A^b uncontroversially ceases to belong to the β-family and M and C remain our only witnesses of the β-text.

F.'s edition makes considerable progress also in so far as the indirect tradition is concerned. (3) Although Ross and Jaeger were aware that for Books Ε–Ν the commentary on the Metaphysics edited by Hayduck in CAG I is not genuine Alexander, they still made substantial use of it, possibly on the assumption that it must have been older than our best manuscripts. The attribution of Pseudo-Alexander on the Metaphysics to the Byzantine scholar Michael of Ephesus radically changes the picture and considerably diminishes the importance of Pseudo-Alexander's testimonies when compared with our best manuscripts. F.'s edition takes full account of the consequence of this discovery. Similarly, William of Moerbeke's translation disappears from F.'s critical apparatus, as it should. (4) F. also makes intelligent use of the Arabic translations of the Metaphysics by Ustāth (ninth century) and Mattā ibn Yûnus (tenth century). Although these translations must be used with some caution, as F. warns us (pp. 152–4), especially when it comes to the minute details of the text and the word order, they cannot be completely disregarded on account of their antiquity (they are older than our oldest and best manuscripts).

For these reasons, F.'s edition of Book Lambda should be regarded as the best edition of the book to date and should certainly replace its predecessors.