It is an enviable task to review Marcus Deufert’s critical edition of the de rerum natura, a poem which, in the current disruptions of our rationes uitae, has taken on special meaning for aegri mortales. The culmination of decades-long, painstaking efforts devoted to the transmission, textual criticism and interpretation of the text, Deufert’s Lucretius now takes its place as the most authoritative and up-to-date critical edition of the poem; it offers a thorough revision of the text on all levels, punctuation, paragraphing, orthography, and wording; and the apparatus criticus, carefully constructed and accurate, tells the fascinating story of how the acts of transcription, correction, interpolation, annotation, collation, recension, emendation, and sheer guesswork, embedded in manuscripts or printed editions or scholarly journals, and enacted by generations of readers, have collectively sustained the energetic and contentious discourse that seeks to recover and verify the poet’s words. The reconstruction and improvement of the transmitted text are based on a meticulously researched history of the transmission and on an impressively wide-ranging review of the textual scholarship. Moreover, a welcome feature, among many, of this edition is that it overcomes the customary limitation of its genre, a brevitas, often obscure to many of its users, in the presentation of historical and textual information. As learned companions that prepared the way for the present volume, the editor published three major studies dealing with the history of the transmission and the criticism of the text. In doing so he has provided students of the poem with an invaluable resource for understanding the historical foundations of the text and the methods by which it has been established, and, as an added dividend, they now have at hand a wealth of material for future research; the Kritischer Kommentar in particular opens a window into the editor’s workshop and allows one to follow his decision-making process. As a result, when readers feel it is necessary to disagree with him, they can do so with a full understanding of how he made his editorial decisions.
The praefatio, written in a clear and concise Latin, covers a number of topics and requires careful reading: the manuscripts and the transmission of the text (pp. vii–xx); printed editions (xx–xxi); an explanation of the three types of apparatus that accompany the text (the critical apparatus for recension, correction and emendation, the apparatus of repeated lines, and the apparatus of fontes from the age of Augustus to the end of the ninth century (pp. xxi–xxix); subscriptiones and tituli (pp. xxix–xxxi; these are given on pp. 297–305); orthography (pp. xxii–xxiii; the editor adopts a carefully controlled inconsistency in the belief that this gives a truer picture of Lucretius’s practice than the thoroughly normalized system based on the spelling habits of the first century AD (e.g., linquontur/lincuntur, rursum/rusum); the fragments, i.e., lines or words absent from the manuscripts but attributed to the poet by grammarians (pp. xxxiii; the texts are given on pp. 306–314). More information on all of these topics will be found in the editor’s Prolegomena, where they are treated in exhaustive detail. In the face of so much information the editor has managed to maintain his sense of humor. To the challenge of the Sphinx-like riddle of distinguishing between the hands of the scribes who copied O and the hands of near-contemporary or slightly later revisers, he responds with modest recognition of what can be accomplished: nam Davus sum non Oedipus (p. ix; see also p. xi on stercus in the apparatus). The praefatio is followed by a list of editions and over 200 secondary works (pp. xxxv–xlv). There is a full-page stemma on p. xlvi, and on pp. xlvii–xlix the manuscripts and their sigla are listed.
In the analysis of the transmission the editor’s most important conclusion is that the lost manuscript of the de rerum natura, commonly referred to as the Poggianus, which Poggio Bracciolini hired a scribe to copy from an exemplar at a German monastery in 1417 and then sent to his friend Niccolò Niccolì in Florence, is a copy of the Oblongus, either directly, the line of descent which Deufert seems to favor, or at one remove, a possibility he recognizes. It is therefore a codex descriptus and does not preserve a stream of transmission independent of the authoritative witnesses, Oblongus O and Quadratus Q, and the three separate gatherings of leaves, the schedae, Gottorpienses G, Vindobonenses priores V, and Vindobonenses posteriores U; G, V and U are now regarded as belonging to the same manuscript. The controversy over the stemmatic significance of the Poggianus (independent witness or copy of O?), can now, in my view, be laid to rest. Readers who are familiar with the somewhat heated debate over the place of the fifteenth-century Italian manuscripts on the stemma codicum will note that only in three places does the siglum π appear for the lost Poggianus, once in the praefatio (p. xv), once on the stemma (p. xlvi), and once in the conspectus siglorum (p. xlviii); it is completely absent from the apparatus criticus. Because π’s importance lies only in the corrections and conjectures made in it by Italian humanists, which were then inherited by the manuscripts copied from π, Deufert made the momentous decision to use the siglum ξ to represent not the lost manuscript π itself but the body of corrections and conjectures written in π, i.e., π’s non-O readings, which are the product of humanist philological activity on the text and have no independent manuscript authority. In this way π’s extraordinary historical significance as the transmitter of the de rerum natura to Renaissance Italy is completely subordinated to its extraordinary significance as the site of an intense textual-critical discourse devoted to the recovery of the poet’s aurea dicta. Although he does not explicitly address the matter in the praefatio, the editor clearly does not accept the claim that papyrus fragments from Herculaneum have been correctly identified as fragments of the de rerum natura. The mystique of the historic triad, O, Q and the schedae, as the foundation of the text remains undiminished.
It will take some time to evaluate in detail the full extent of the editor’s contribution to the revision of the text and the apparatus criticus; his textual decisions and conjectures will provoke discussion for years to come, a clear sign of the edition’s importance. Overall, his judgment, informed by a deep knowledge of the manuscript tradition and centuries of textual scholarship, proves to be a steady, reliable and informative guide in the establishment of the text. I find an engaging boldness in his approach. For example, at 3.58 he does not accept the old standby of the print tradition, the highly problematic manet res, a humanist conjecture, and instead obelizes the reading of the archetype †manare†. This will bring renewed attention to the problem. On the other hand, in a case where readers might expect the obelus, the locus desperatus at 4.545, et †ualidis nete (necti O1) tortis† ex Heliconis Ω, Deufert prints his restoration of the line, which is essentially a refinement of conjectures proposed by others: et uolucres uiridis nocte oris ex Heliconis. Although it may not be what Lucretius wrote, it deserves serious attention as a compelling reinterpretation of the ductus litterarum. Of new, or less commonly printed, emendations in the text readers will want to note the following: 1.469 Teucris (Munro); 473 amoris (Wakefield); 843 de parte (Diels); 887 lanigeris (Butterfield); 2.88 ad tergus ibi (Courtney); 181 quae quantast (Deufert); 219 deflectere (Watt); 474 nam fit dulcis (Deufert); 547 hoc tibi si sumas (Deufert); 1034 uisunda(Diels); 3.420 mente Müller; 444 incohibentist (Eden); 493 spumas animam (Zwierlein); 694 subsit si Clark; 820 fatalibus(Zwierlein); 962 iuueni (Deufert); 4.553 una Bentley; 875 toto (Deufert); 1026 parui (Clarke); 1096 mentem spe raptant (Müller); 5.514 inferni (Deufert); 1353 insubla (Deufert); 6.778 tractu (Polle); 1012 quo ducitur ex elementis Lachamnn; 1064 Magnesia flumina saxi (Bentley). This is just a sample.
Of course, there will be disagreements about editorial choices. Here follow some observations; the reading in Deufert’s edition is given first. 1.5 … solis: The complacent colon at the end of 1.5 arrests the syntactic movement of the invocation; Vahlen’s punctuation of 1.1–28 as a single, exquisitely articulated period should be adopted. 1.50 <Memmi>, uacuas auris <animumque>. The attractions of supplying a vocative here, with Sauppe’s supplements, must yield to the hard evidence of the quotation, uacuas auris animumque sagacem, attributed to Lucretius, in the schol. Veron. on Vergil, Geo. 3.3, and placed in 1.50 by Bernays. 1.66: tendere Nonius] tollere Ω. tollere should be retained; it is a constructive element in the poet’s polemic against the traditional teleological argument that the upright posture of human beings is purposeful, allowing them, unlike other animals, to observe the firmament, perceive its order and conclude that there is a divine power that governs that order. Epicurus raises his eyes to the heavens, not to fulfill a misguided concept of our species’ teleology but to reveal the nature of the universe and thus enlighten his fellow humans (fearfully prone, iaceret1.62) about their own nature. 1.105 uitae rationem euertere Deufert] uitae rationes vertere Ω. The editor assumes, incorrectly in my view, that the poet must be speaking of ‘die rechte Lebenswiese’. At this stage in the poem Memmius is not yet in possession of the uera ratio. Clearly, tuas in the following line is to be understood with rationes ‘your life-plans’, which, for the as yet unenlightened Memmius, can easily be upended.
1.161–162 e caelo … terra Deufert] e terra … caelo Ω. It is difficult to understand the editor’s reason for transposing terra and caelo. uolucres erumpere terra is not a shocking adynaton, especially since birds had their origin in mater terra. 2.1 The words suabe [sic] mari magno, inscribed on the wall of a Pompeian house, should be included among the fontes. 2.462 laedens Deufert sed rarum Lachmann] sensibus sedatum Ω. Deufert and Lachmann have teamed up for a substantial rewriting of the transmitted text. I prefer a gentler remedy; for sedatum read serratum (‘whatever is sharp-toothed to the senses’; cf. 2.432 dentata and 463, acutis…elementis). 3.58 <et> (ALo) eripitur Ω. Brieger’s conjecture deripitur is worth including in the apparatus; it is the right word for pulling or snatching something off the body and it preserves the asyndeton. To the objection that it is not consistent with the poet’s metrical practice (see Munro, vol. 2, pp.13–14), perhaps an exception may be allowed on the evidence of 3.527 and 4.554. 3.657 minanti Ω] micantiLachmann. The movement of the snake’s tail is most accurately described by micanti. 3.993 †uolucres† Ω, very likely repeated from 984 or 880, as the editor notes. The general word for torments needed here, which is consistent with the verb lacerant, is stimuli; read quem stimuli lacerant. 4.582 montiuago Deufert] noctiuago Ω. The editor’s montiuagocorrectly identifies the traditional habitat of fauni. The scene, however, has already been clearly set in 4.572–579. Essential here are the circumstances in which the mind is prone to false judgments about sense perceptions, i.e., hearing sounds at night in isolated locations, somewhat similar to the experience of the dreamer described at 4.453–461 (severa silentia noctis 460). Faunus himself speaks at night (Vergil, Aen. 7.81–103; Ovid, Fasti 4.662–670; cf. Statius, Theb. 4.695–696). 5.1142 †propter odores† (= 2.417) Ω. It is disappointing not to see in the text Housman’s convincing restoration of the line, iam maris ueliuolis florebat nauibus pontus, which is based on Servius’s quotation, on Aen. 7.804, of the phrase florebat nauibus pontus from Lucretius. I also miss a reference to S. Timpanaro’s inspiring defense of Housman’s proposal. 6.1263 confectos Susius] confertos Ω. confectos is an odd choice in a context where the poet is emphasizing that overcrowding in the city (confertos) increased the contagion. In a poem that contains the whole shebang, summa summai totius omnis, there is plenty of room for debate.
These are exciting times to be a Lucretian. And this major contribution to Lucretian and Latin scholarship is sure to generate even more excitement. One hundred seventy years after Lachmann’s edition it seems fair to say that much has been accomplished and much remains to be done.
 Pseudo-Lukrezisches im Lukrez: Die unechten verse in Lukrezens “De Rerum Natura,” Berlin/New York 1996; Prolegomena zur Editio Teubneriana des Lukrez, Berlin; Boston 2017 (reviewed by Nicoletta Bruno, BMCR 2019.05.17); Kritischer Kommentar zu Lukrezens De rerum natura, Berlin; Boston 2018 (reviewed by Nicoletta Bruno, BMCR 2019.06.18).
 For full details see the editor’s Prolegomena or D. Butterfield’s The Early Textual History of Lucretius’ De rerum natura, Cambridge 2013 (reviewed by Lisa Piazzi, BMCR 2014.10.21). Michael Reeve must be named here for his work on the Italian tradition, truly a rerum inuentor.
 For a different view of the stemmatic importance of the lost Poggianus and detailed reports of the readings in the 15th-century manuscripts see E. Flores, ed., Titus Lucretius Carus. De rerum natura, 3 vols., Naples 2003–2009.
 R. R. Benefiel, “Dialogues of Ancient Graffiti in the House of Maius Castricius in Pompeii.” AJA 114 (2010): 68 and 93, Cat. no. 37
 Hermes 138 (2010): 67–69.
 Contributi, Rome 1978: 146–193.