This revised Cambridge Ph.D. thesis, written under the supervision of Michael Reeve, is primarily a study of manuscripts O (Leiden, Voss. Lat. F. 30) and Q (Leiden, Voss. Lat. Q 94), on which our knowledge of the text of Lucretius depends. It is precise, meticulous and will form the basis of any future editorial work on the text, including the Oxford Classical Text of Lucretius which B. is preparing. From B.’s study there emerges a concrete list of recommendations on how a critical apparatus should be constructed, notably which corrections in the manuscripts it ought to report. B. provides definitive refutations of a number of troubling minority views regarding aspects of the transmission and provides clarification and articulation for numerous majority views (cf. the earlier summary account of the transmission by L.D. Reynolds in Texts and Transmission. A Survey of the Latin Classics (1983), pp. 218–22). As a demonstration of how the close study of a couple of manuscripts can enable scholars to reach back deeply into centuries of the history of a text, this book is a model of its kind. It does in no way diminish B.’s achievement that his findings are consistent with M. Deufert's observation that improvement of the text of Lucretius will come from careful examinatio (M. Deufert, ‘Zu den gegenwärtigen Aufgaben der Lukrezkritik’, Hermes 138 [2010], 48–69, at 49).
The first chapter gives an overview of the extant manuscripts. B. proposes (p. 10) to use the siglum ‘S’ for the so-called Schedae Gottorpienses, Schedae Vindobonenses priores, and Schedae Vindobonenses posteriores, since it is sufficiently likely that these are fragments of the same manuscript (which descends from the same hyparchetype ψ as Q). There follows an account of editorial progress since the nineteenth century and a scholarly doxography on the Italian tradition of Lucretius. Anticipating the findings of his Chapter 4, B. confirms that all the Italian manuscripts ultimately go back to O via a descendant of O which he dubs χ, but the picture is complicated by the fact that humanist readers occasionally, in a desire to improve the text, compared χ directly or O after it had been corrected further, i.e. after it served as the exemplar for χ.
The second chapter deals with the indirect tradition of Lucretius, strictly with a view to assistance it might provide to the constitution of the text. No author before Seneca the Elder quotes Lucretius, although there are of course allusions to his poem in a number of earlier authors. Lactantius probably read a text which diverged from that of the extant manuscripts early (p. 59). Nonius Marcellus preserves the correct reading against the direct tradition in two places (p. 67), 1.306 and 6.1195. Macrobius had access to a less corrupt text than Nonius (p. 69), but is likely to have drawn on works of Virgilian scholarship, so that no general view on his text of Lucretius can be taken. Isidore's use of Lucretius had not received much attention, and he quotes lines of Lucretius which nobody else quotes. Further, there is no reason to think that he ever left Spain. So his evidence is likely to be original and tells us something about the circulation of Lucretius in Late Antiquity. Overall the indirect tradition can provide help or at least worthwhile material for the apparatus in only a few places (p. 101). The section concludes with a discussion of the so-called fragments of Lucretius, lines ascribed to him in various sources but absent from the direct tradition. Only three of the sixteen fragments are judged by B. to be conceivably genuine, and none confidently so (p. 135); finding possible locations for them in the transmitted text is difficult.
The third chapter is the most comprehensive study to date of the capitula to be found in the manuscripts of Lucretius, section headings introduced into the body of the text as well as collected at the beginning of books so as to form a kind of table of contents (indices). These capitula offer fascinating insights into the transmission of the text and the way in which it was studied. They help with the constitution of the text only in places where it was corrupted after the capitula were derived from it (4.526, 6.890, possibly 4.216–29). B. offers a detailed chronology for the emergence of the capitula and their assembly into indices (summarised pp. 201–2), which advances beyond earlier work by clarifying many points of detail (see, e.g., pp. 150 n. 39, 162). He surveys the form of the Latin capitula, concluding that they arise from the same individual (p. 185) and describes the mechanisms by which capitula were derived from the text (ruling out the implausible view that similarities in phraseology between text and capitula point to Lucretius as the author of the latter). He ends with the sensible recommendation that future editors of Lucretius should assemble capitula and indices in an appendix, mentioning them in the apparatus only where they have a bearing on the text.
The fourth chapter deals with the correcting hands of O. While Q is largely free of corrections, O had a number of readers who made corrections or left glosses, and any editor will need to tell them apart as far as possible, so as to be able to judge an individual correction in the context of other efforts by the same person, and to be able to distinguish readings with a chance of being informed by manuscript evidence. B. has done this more carefully than anyone before him, has described the layers which can be discerned and lists his findings for each of the hands distinguished in Appendix 3 as well as, in more detail, on his website (p. 204 n. 5).Footnote 1 Apart from presumably instant corrections of the original scribe (listed p. 203 n. 2), the most significant layer of corrections is due to an early ninth-century insular hand which Bernhard Bischoff identified as the Irish scholar-monk Dungal, whose work receives the fullest documentation and analysis to date. Dungal supplied lines which had been omitted by O but are in Q(S), and thus evidently used a manuscript for comparison. This manuscript, B. holds (like Reeve, n. 3 p. 159), was Ω, i.e. O's exemplar (p. 214). However, Dungal also made emendations ope ingenii, and B. discusses the question of how to distinguish these from likely readings of Ω which have been corrupted in the other branch of the tradition (i.e. Q and its source ψ). Apart from their narrow purpose, such observations add to our overall picture of the intellectual life of the Carolingian period.
The book concludes with a series of appendices. Appendix 3 was already referred to above, others are devoted to the capitula, to the authors from Seneca the Elder to the end of the Carolingian period who cite Lucretius, to the correcting hands in O., the foliation of the archetype of Lucretius, and the fate of OQS in the early modern period.
