Philodemus, On Poems Book 5
David Armstrong from the edition of
Cecilia Mangoni, Naples (Bibliopolis), 1993
Mangoni's text--as I have verified by reading the sources ( P.Herc. 1425 and 1538 and the N and O disegni of them) in Naples and Oxford in summer 1993--is an immense improvement on the previous text, that of C. Jensen ( Berlin, 1923). I did this work as part of a NEH Translation Grant, directed by Richard Janko, David Blank, and Dirk Obbink, which will result in a text with English translation and notes of On Poems, On Music, and On Rhetoric. So much speculative Greek of Jensen's has been removed and replaced by Mangoni with either excellent Greek of Philodemus' own or honestly admitted lacunae that the effect is like that of a darkened monument or painting newly and splendidly cleaned. Where I differ from her text or interpretation I have indicated this in a note. The headings I have added indicate (a) current opinion, as represented in Mangoni's preface, on the structure of the surviving part of Book 5, and (b) my own personal opinion--which is more adventurous in some respects than Mangoni's--of what positive axioms about poetics of Philodemus' own can be gleaned from his criticism of other critics.
What Jensen, whose fragment and column numbers are those of N, the Naples transcript) numbers as "fr." i and ii are numbered by Mangoni cols. i and iii. In Jensen and N there then follow "columns" i-xxxvi. The disegnatore of O, the Oxford transcript, numbered these "fragments" cols. i and ii, resulting in 38 columns in all. Mangoni agrees with him but has added what can be deciphered with the aid of the microscope (a little) of the column that came in between the two, now numbered in her edition col. ii. The result is that her column numbers are (after iii) those of Jensen and N +3 (e.g. Jensen's and N's col. i = Mangoni col. iv, their col. xii = Mangoni col. xv). With relation to O, they are O +1 (O col. iii = Mangoni col. iv and so on up to the final column, O xxxviii = Jensen and N xxxvi = Mangoni xxxix). For the convenience of readers comparing Jensen's edition, which follows the numbering of N but also gives (inadequate) reproductions of O, I list all three.