Horace has been well served by the ‘Green and Yellow’ series, Cambridge Greek and Latin Classics: the appearance of this volume leaves only Satires Book 2 and Odes Book 3 without commentaries, and A. J. Woodman is at work on a commentary that will fill the latter of those gaps. Horace has also been well served by Stephen Harrison: his bibliography lists seventeen Horatian contributions ranging over three decades. With hindsight it seems almost inevitable that H. would eventually produce a commentary on Horace, which he has now done with highly satisfying results.
A crisply written Introduction covers much ground in just over twenty pages. Topics discussed include date (on which H. shows some sympathy for Gregory Hutchinson's argument that the three books of Odes were published separately before being collated in c. 23 b.c.e.), Horace's literary career, characteristics of Odes 2 (moderation and consistency following the exuberant variety of Odes 1, a focus on friendship and ethical questions), arrangement (H. sensibly points to groupings of poems rather than attempting to demonstrate an overarching structure for the whole book), literary intertexts with emphasis on allusions to the Georgics and to the Orpheus story in particular, structural elements (especially ring-composition, a ‘turn’ in the middle of an ode and favourite closural devices), style (brief remarks on diction and word order followed by a stylistic analysis of 2.14), metre and text.
H. has established his own text; it contains some novelties that will be discussed below. He presents the poems with no breaks between stanzas, on the grounds that ‘this is how they appear in our earliest Greek papyri’ (23). A minimal apparatus registers only the most significant variants and conjectures.
The lemmatised commentary on each ode is prefaced by a summary of its contents, an indication of metre, an introductory essay and a select bibliography. As H. himself notes in his Preface, any commentary on Odes 2 will be to some extent indebted to the 1978 work of Robin Nisbet and Margaret Hubbard (hereafter N–H). It is to H.’s great credit that his work does not suffer from the unavoidable comparisons with its eminent predecessor. He does not seek to imitate N–H's generous citation of parallels, instead often referring his readers to N–H for fuller discussion (thirty times in the notes to 2.16 alone). But within his more restricted space he has found room for many acute observations not found in N–H, of which only a handful can be mentioned here. 2.1.35 ‘decolorare is a rare and prosaic verb (only here in Latin poetry)’. H. is consistently alert to issues of diction and register. 2.2.8 Fama superstes ‘Fama will keep Proculeius’ reputation alive after his death, behaving like a surviving relative’. 2.5.9 colloquial force of tolle (‘get rid of’). 2.7.15–16 echoes of sea storms in the Odyssey could create ‘a neat Odyssean balance to Horace's Iliadic escape’. 2.11.12 H. cites an apt Lucretian parallel for animum fatigare. 2.14.5–7 ‘The theme of the insufficiency of lavish bovine sacrifice to avoid death recalls Il. 22.168–71’. The introductory essay to 2.17 adduces Catullus 11 as an important intertext, and the essay on 2.19 brings out a similarity between Bacchus and Horace (in 3.1) as teachers. (One might add, with reference to 2.20, that both have metamorphic powers.)
When H. parts company with N–H, he not seldom has the better of the argument. 2.14.28 H. must be right to take potiore as a genuine comparative (‘outstripping the dinners’) rather than ‘as desirable as a dinner’ (N–H). 2.16.17 H. is again right in taking iaculamur as ‘drawing in the reader as a fellow mortal’ rather than as a ‘bedside plural’ (‘how are we today?’ N–H). 2.18 H. takes the tu of 17 as the generalised second person of diatribe; that seems more likely than N–H's attempt to see as well a reference to Maecenas (difficult after he has been clearly alluded to in the third person just before, 12–13 ‘nec potentem amicum largiora flagito’). 2.18.23 usque is more likely to mean ‘continually’ (H.) than ‘progressively’ (N–H).
H. also has the benefit of four decades of Horatian scholarship since N–H, and he has fully incorporated recent work into his notes.
One feature of N–H that is conspicuously absent in H.’s commentary is the capsule judgement of an ode, often functioning as the conclusion to the poem's introductory essay. (On 2.1: ‘here for once < Horace > speaks not simply as a panegyrist but with the judiciousness and authority of a historian’; on 2.2 ‘once again he has surprised us with a brusque and dry but brilliantly concentrated and most original poem’.) H. may have avoided statements of this kind as being too subjective for a commentator, but the result is a diminished sense of a personal voice: one cannot easily tell which poems H. especially likes or regards as particularly successful.
Any commentary that puts forth fresh ideas will provoke occasional dissent, and I am not persuaded by some of H.’s textual and interpretative moves. 2.1 the suggestion that 17–24 describe Pollio's history in terms that resemble a performance of his stage work is not attractive. 2.1.20 H. does not make a strong enough case against the transmitted uultus to justify his proposed emendation to pectus. 2.5.13 H. suggests taking ferox as the vocative of a proper name (citing C. Iulius Ferox, consul of 99 c.e.); but would Horace have placed a name in a position where it could easily be misread as an adjective, and indeed how could a reader of a text without punctuation or capitalisation have realised that ferox was a proper name? 2.6.7 H. takes lasso as referring to Septimius rather than Horace; after a line explicitly relating to Horace (‘sit meae sedes utinam senectae’), one might expect a change of referent to be signalled. 2.9.18 H.’s note says that he marks a comma after noua, treating it as a neuter plural noun rather than adjective with tropaea (again, how would an ancient reader have known?), but his text has no comma. 2.12.9 for the admittedly difficult tuque H. proposes tu ipse; tu is not elided elsewhere in the Odes and tu ipse seems harsher than the tu in numerum of Epist. 1.14.41. 2.12 H. believes that Licymnia is a hetaira with whom Horace is involved. Other difficulties apart, Horace cannot easily assert Licymnia's fidelity to him (15–16 ‘bene mutuis / fidum pectus amoribus’) and also remind his reader (as H. says on l. 23 permutare) ‘that Licymnia's services are for hire’. H. thinks that 9–12 may be suggesting that Maecenas produce a prose version of the triumphal passage on the shield of Aeneas, ‘possibly written by this date and known to both <Horace> and Maecenas before publication’. Since H. also suggests that the use of ‘Caesar’ rather than ‘Augustus’ implies a date of composition before 27 b.c.e., that puts the drafting of the Aeneid implausibly early. 2.12.9 H. takes pedestribus as denoting ‘plodding prose’; would it not be tactless for Horace so to characterise the history he is suggesting Maecenas might write? 2.14.25 H. adopts Campbell's degener for the transmitted dignior, replacing an interestingly paradoxical reading with a blandly conventional one. 2.19.31 H. prints his own conjecture cauda for caudam, which simplifies the grammar and is worth considering, although he perhaps exaggerates the difficulty of the transmitted reading and introduces an elision not found elsewhere in the Odes (only in Epodes and Satires).
It has been said that the best commentaries reflect the qualities of the texts on which they are based. H.’s work may not possess all the Horatian virtues — wit and irony are in short supply — but it certainly displays the moderation and consistency that he defines as hallmarks of this book of the Odes.
