1. Introduction
Carl Becker's 1963 book defined all these works plus the first book of Epistles as constituting Horace's late work, partly because (in the biographizing fashion of the time) he saw Epistles 1 as the decisive beginning of a final and mature period for the poet, focused on philosophical and ethical retirement and distanced contemplation of poetry and the world. In this volume I have chosen to assign the first book of Epistles not to the later period but to the middle period with the Odes (see Chapter IV), partly because I hold that it is not so different from the Odes in its concerns and techniques, even if it constitutes a move from lyric back to the hexameter sermo which Horace had last used ten years before, and partly because I take the philosophical programme of Epistles 1 as a statement about the book's particular content rather than about the poet's life in general.
Becker dated Epistles 2.2 to 19 bce and the Ars poetica to 18 bce. The latter at least is not much favoured by modern scholarship: most are now moving to the idea that the Ars is a final massive statement on poetry capping the views of Epistles 2, perhaps even originally intended as part of that book – see Chapter II above. However, it seems clear in general terms that these poems of Horace's last decade are written from the perspective of a senior poet, perhaps the last of the great Augustan generation, following Virgil's death c.19 bce and the likely silence of Propertius after his final book of elegies c.16 bce. In this period national/political interests and (especially) literary didacticism predominate over more ‘frivolous’ themes, which are effectively rounded off in the fourth book of Odes.
2. Carmen saeculare
After languishing for some years in the shadows, the Carmen saeculare has in the last generation emerged into the full light of scholarly scrutiny.Footnote 1 Critics have begun to realize that this seventy-six-line poem, though composed in the Sapphic stanza, the second most popular metre in the Odes, is no mere appendage but fundamentally distinctive within Horatian lyric.Footnote 2 It was written (as an inscription confirmsFootnote 3 ) in the first-person plural for performance by a choir of twenty-seven boys and twenty-seven girls in the ritual context of the ludi saeculares at Rome on 3 June 17 bce.Footnote 4 This presents a major contrast with the usual single voice of the rest of Horatian lyric (see Chapter IV). Modern scholarship has been much concerned with working out how the text of the poem relates to its religious context,Footnote 5 but has also provided in-depth readings of the poem as a literary artefact,Footnote 6 and explorations of its affinities within performed Greek lyric genres such as the paean.Footnote 7 The fact that this is the one poem of the Augustan period for which we have firm external evidence for a performance context has naturally moved it to the centre of the recent debate about performance elements in Roman poetry.Footnote 8 In terms of style and structure, its somewhat prosaic language can be linked to its ritual context, with its need to point to particular divinities and features of their cults, while its triadic structure (six groups of three stanzas each plus a final single epode) plainly looks back to Greek choral lyric.
3. Odes 4
The claim of the ‘Suetonian’ biography of Horace (see Chapter II) that the fourth book of Odes grew from Augustus' commissioning of Odes 4.4 and 4.14 for the two princes Tiberius and Drusus, which stimulated a return to lyric odes after a long interval, looks likely to be a conjecture from the prominent presence of these poems in the book. It seems more probable that Horace's return to a different type of lyric through the commission of the Carmen saeculare of 17 bce led to more lyric poems in the manner of Odes 1–3, and, as critics point out, memories of the Carmen saeculare and its dramatic publicity for the poet are a feature of Odes 4 (4.3, 4.6). Though political poems predominate in this book, there are also some poems in the manner of Horace's earlier collection, especially in its later part (4.1, 4.10, 4.11–13). In terms of dating, although some had viewed Odes 4 as Horace's latest work,Footnote 9 most would now see the book as written in the period 17–13 bce and emerging in 13 as Augustus returned from Spain and Gaul (prefigured in 4.2 and 4.5).Footnote 10 The publication of a literary work as an anticipatory celebration of an Augustan return would be paralleled by the probably appearance of Virgil's Georgics in the summer of 29 bce, in time for the triple triumph of August of that year.Footnote 11
The fourth book, though given detailed treatment by Fraenkel as the climax of Horace's career,Footnote 12 was not especially popular in the 1960s and 1970s, perhaps because of the predominant political and encomiastic material, which did not fit the times especially well.Footnote 13 Once again, Michael Putnam deserves some credit for rescuing a Horatian book: his running commentary (Putnam 1986) was an important stimulus to further work, and is still the most extensive literary treatment, clearly showing that the poetic texture of the fourth book matches those of the earlier ones in complexity and interest. The same is true of the de facto running commentary of T. Johnson 2004, which neatly argues that Book 4 effectively combines symposiastic and encomiastic elements in drawing the audience from private to public celebration, while maintaining a Callimachean witty independence and ambiguity.
The lack of a focused modern conventional commentary outside those on the whole of the Odes by Quinn (1980) and Syndikus (1972–3, second edition 2001) was also certainly something of a barrier to the book's study for some years. However, we now have an extensive Italian commentary by Fedeli and Ciccarelli (2008), which has a good deal of important material, especially on language, and a major commentary in English by Richard Thomas for Cambridge (2011), which applies with excellent results its author's characteristic emphasis on poetic texture and anxieties about encomiastic elements in Augustan poetry.
The increased encomiastic element in Book 4 is indeed striking. It has been plausibly linked to an increased interest in Pindar, named in 4.2 as inimitable and 4.9 as excellent, and clearly a model for the praise odes of this book. These use Pindaric techniques of victory laudation such as extensive heroic speeches and gnomic utterances, applied here not to Pindaric athletic success but to the parallel triumph of Roman arms.Footnote 14 Another significant element is the clear presence among the addressees of the fourth book of a generation of young men rising to prominence under the middle years of the Augustan regime and on the verge of the consulship as Horace writes.Footnote 15 Odes 4.1 is addressed to Paulus Fabius Maximus, husband (then or later) of Augustus' cousin Marcia and soon to be consul in 11 bce; 4.2 to Iullus Antonius, son of Antony, brought up in the imperial house and soon to be consul in 10 bce; 4.4 to the prince Drusus, stepson of Augustus, consul 9 bce; 4.14 to his brother, the future emperor Tiberius, consul 13 bce; 4.8 (perhaps) to C. Marcius Censorinus, consul 8 bce. The princeps himself is directly addressed in 4.5 and 4.15, really for the first time by Horace in his Odes (1.2 and 1.12 had praised him, the Roman Odes had evoked him at a distance, and 3.14 had welcomed him back in the third person). Maecenas is restricted to one (warm) mention in fifteen odes (4.11.19–20), perhaps indicating continuing friendship but a naturally diminished role as the poet's channel to the princeps, who was now personally connected with Horace after the Carmen saeculare and more available to him in Rome in the teens bce.Footnote 16
Alongside this more political element sits erotic and sympotic material recognizably akin to that of the first three books. The book begins (4.1) with a supposed resurgence of erotic feelings in the poet, this time unrequited passion for the boy Ligurinus, also the addressee of an epigram-style poem which warns him against unkindness to lovers, given that he will lose his looks in time (4.10). These poems surely point symbolically to the reprise of erotic lyric rather than to a real obsessive passion, and to the potential inappropriateness of erotic poetry for an ageing poet past fifty (4.1.4–6). The same message emerges from 4.11, inviting Phyllis to a party, with the claim that she is his last female lover, readily interpreted as a metapoetical statement that the poet is now coming to the end of his lyric output. In 4.13 the poet rejoices that Lyce has become old but also laments his own ageing, while the famous spring ode (4.7) greets the cyclic return of the seasons but also looks forward to the final stage of the human life cycle in death. This material is self-consciously the work of a poet who has passed the youth appropriate to love and is looking to the end of his lyric career: the opening presentation of Venus as goddess of love in 4.1 is balanced and capped by her role as national goddess in 4.15.Footnote 17
An interesting problem is set by 4.12, addressed to a Vergilius and replete with allusions to the poetry of Virgil, but this Virgil is alive and socializing with young men, whereas the great poet has been dead for several years by Odes 4. Is this a poem with a past dramatic date, or is this Vergilius not the poet? Various views have been expressed:Footnote 18 one possibility not yet adequately discussed is that this could be a living relative of the dead poet (hence the Vergilian allusions, but an addressee still alive after 19 bce). It would certainly be unparalleled for a Horatian ode to be addressed to a dead person as if he were alive.
4. Epistles 2 and Ars poetica
Following discussions on dating since Becker 1963, scholarship seems to be moving towards a consensus that both Epistles 2.1 and the Ars poetica belong to the very last phase of Horace's career, between Odes 4 (c.13 bce) and Horace's death in 8 bce. A more disputed issue has been the dating of Epistles 2.2: its sharing of an addressee (Florus) with Epistles 1.3 and its apparent reference to the same military service with Tiberius there mentioned (2.2.1~1.3.1–2) have usually led to a dating about the same time as Epistles 1 (19 bce), but this leaves a poetic book with two items, one of which (the second) is some seven years earlier than the other. A recent development has been the argument that the reference to Florus' faithful service points to his being with Tiberius since the previous poem some years before: the poet thus praises his friend's laudable loyalty to his commander over a considerable time.Footnote 19 This leaves us with the possibility that both Epistles 2.1 and Epistles 2.2 can belong to the same period, c.12–8 bce.
The date of the Ars poetica seems to have settled down after a period of some debate.Footnote 20 Williams 1972, agreeing roughly with Becker 1963, had argued for an early date of 23–17 bce, and a date of 23–20 was proposed following stylometric affinity with Epistles 1 by Frischer 1991, but most have now accepted Syme's arguments about the identity of the addressees as the sons of Piso the Pontifex, and the consequent need for the poem to be published after Piso's own absence from Rome in 12–10 bce.Footnote 21 Caution pinning down a clear date for the Ars is still expressed by Brink,Footnote 22 but he too is generally inclined to date it after Odes 4.Footnote 23 Williams had already suggested that, given its epistolary form and literary concerns, the Ars poetica would naturally form a Horatian poetry book of normal length with Epistles 2.1 and 2.2 (a feeling shared by Kilpatrick), which he would have dated to c.17 bce.Footnote 24 Harrison 2008 has taken these ideas further, arguing that the Ars poetica was conceived as the climax of Epistles 2, though its apparent early separation from the rest of the book in the transmission and indirect tradition (it is cited as Ars poetica as early as Quintilian 8.3.60) might suggest that it was published posthumously and separately. The idea that it might originally have been intended as Epistles 2.3 has been regularly considered by scholars,Footnote 25 and there seems little doubt in general that the Ars poetica is closely bound to Epistles 2.1 and 2.2 through its epistolary framework and literary-ethical didactism.
This unity is underlined in the massive enterprise of Brink's Horace on Poetry, which in three volumes over a period of several decades (Brink 1963, 1971, 1982) produced texts, commentaries, and literary considerations of all three poems, Epistles 2.1 and 2.2, and the Ars poetica. Of these, Epistles 2.1, with its address to Augustus, had received a substantial treatment in Fraenkel 1957, but the other two poems were conspicuous by their absence from that book, which left Brink ample room for manoeuvre. In many ways, Brink's work here is parallel to that of Nisbet and Hubbard on the Odes, bringing to bear a huge panoply of scholarship in the exegesis and interpretation of these difficult poems, and constituting a monumental achievement. Since (unlike Nisbet and Hubbard) it produces a critical text with apparatus, and there are more textual difficulties than in the Odes, textual discussion looms much larger, and Brink's discussion of the Horatian transmission has remained authoritative, while his expertise in ancient philosophy and Latin lexicography is clear throughout.Footnote 26 Notes in the commentaries are generous in length and cover a full range of learning from textual and linguistic to cultural and philosophical, and they are sometimes hard to navigate; the 962 lines of these three poems receive (in addition to a substantial volume of general introduction) some 1107 pages of exegesis, more than twice the density of Nisbet and Hubbard on Odes 1 and 2 (775 pages on 1436 lines). This massive commentary, already a generation old, will not be soon superseded in its philological and philosophical discussions, though its bibliography (always somewhat backward-looking) is naturally becoming dated, and its literary-critical interpretations can be debated and questioned.Footnote 27
As with Nisbet and Hubbard, the existence of a vast major scholarly commentary has stimulated the emergence of smaller-scale works which incorporate and challenge its findings. The commentary of Rudd (1989) has done this admirably for all three works, using Brink circumspectly and independently, and showing itself shrewd and sensible in its interpretations and judgements. Also very useful is Kilpatrick 1990, a fine running commentary on the three poems, which brings out their shared interests and themes clearly and sympathetically (see also Kilpatrick 1986), and adds an elegant translation of all three. Handbook treatments have been similarly effective: Laird 2007 provides a succinct and stimulating account of the Ars with an excellent orientation in modern scholarship, while Günther 2013c gives an extended literary reading of Epistles 2.1 and 2.2, and Reinhardt 2013 argues that the Ars is an effective channel for Peripatetic literary criticism and uses techniques for engaging the reader already established in Latin didactic poetry by Lucretius.
It may be worth pointing out some of the clear thematic similarities which hold the three poems together, and which can be taken as possible evidence for their planned inclusion in a single book.Footnote 28 Both Epistles 2.1 and the Ars focus on the development of Latin literature from Greek, especially with attention to drama (Epist. 2.1.156–67; Ars 268–88): the two passages contain the only mentions in all Horace of the names of Thespis (Epist. 2.1.163; Ars 276) and Aeschylus (Epist. 2.1.163; Ars 279). Both Epistles 2.1 and the Ars contain criticism of Plautus, regarded by Horace as crude and undiscriminating (Epist. 2.1.170–6; Ars 270–4), and of the Alexander-poet Choerilus of Iasos, seen in both poems as an incompetent anti-model for modern epic poets, again mentioned only in these two contexts in all Horace (Epist. 2.1.232–3; Ars 357–8). The prominent pair of modern epic poets Virgil and Varius is likewise found in these two poems (Epist. 2.1.247: Vergilius Variusque poetae; cf. Ars 55: Vergilio Varioque), recalling their pairing in the first book of Satires more than twenty years before (Sat. 1.5.40–1: Plotius et Varius Sinuessae Vergiliusque / occurrunt [‘Plotius and Varius and Virgil met us at Sinuessa’]; Sat. 1.6.54–5: optimus olim / Vergilius, post hunc Varius, dixere quid essem [‘once excellent Virgil, and after him Varius, said what kind of thing I was’]). The double reference is perhaps a career-closing return in a final collection to Horace's first poetry book and a tribute to his now dead poetic friends.
Epistles 2.2 and the Ars can be equally well connected. Both begin with famous counterfactual conditions as opening devices – compare Epistles 2.2.1–3, 17 (‘were someone to try to sell you a slave with a suitable sales patter, he would succeed …’ ) with Ars 1–3, 5 (‘were an artist to add a horse's neck to a human head and bird's feathers … would you not laugh?’). Both poems also contain lists of Horatian genres. At Epistles 2.2.59–60 Horace's friends are said to differ in their preferences between Odes, Epodes, and sermones, while Ars 79–85 treats key elements of the poet's career with elegant indirection, moving from Archilochus (as the model of the Epodes), via a digression about the use of iambics in drama, to a Pindarizing account of lyric which clearly encapsulates the major themes of the Odes: hymns, epinicians (no doubt looking to the Pindaric imitations of Odes 4), love, and the symposium. Once again, the element of surveying the poet's output, whether explicitly or implicitly, would be appropriate to a unified and self-consciously ‘late’ book in Horace's poetic career.
All three poems share key thoughts too. Each deals with the theme of the usefulness of the poet to the community: Epistles 2.1.124 memorably claims that he can be utilis urbi (‘of use to the city’), and Epistles 2.2.121 that the poet will make Latium rich with his language, while Ars 396–401 points to the historic function of poets as law-givers. Typically, at the end of each poem this dignified idea of the poet's status is conjoined with a more ironic presentation of the same idea. Epistles 2.1 concludes with the fate of bad poets and their verses which Horace seeks to avoid (being used to wrap spices: Epist. 2.1.267–70); Epistles 2.2 with a playful self-address which suggests that the poet has enjoyed more than enough of the pleasures of life (Epist. 2.2.213–6); and the Ars with the celebrated picture of the mad poet who will not leave his listener alone (Ars 453–76).
The poet's direct engagement with the princeps in Epistles 2.1, the epistle to Augustus, has attracted particular scholarly attention in recent years, especially in terms of its political subtexts. Oliensis 1998 has shown that the poem tackles issues of the poet's status and ‘face’, pointing to the dangers of literature as gift exchange and noting that Horace leaves himself out of the list of poets personally patronized by Augustus.Footnote 29 Feeney 2002 argues that the literary-critical content of the poem necessarily involves political ideology in matters of freedom of speech, authority, and canon-formation, and that there is a clear analogy between the ageing princeps, ruling alone after the death of Agrippa, and the ageing poet, left alone at the summit of Roman literature after the death of Virgil and others. Lowrie 2009b examines the issue of the social function of Horatian poetry through the epistle's stress on the apparent loss of the Greek immediacy of performance in the written literary culture of Rome, arguing that Horace is conscious of the inevitable politicizing of performance culture and its conflict with the tendency of lyric to identify an individual, non-conformist voice.Footnote 30 The same issues have been detected in Epistles 2.2: Freudenburg 2002 locates the poem in an atmosphere of anxiety about Rome's renewal at the time of the ludi saeculares of 17 bce, detecting consequent political and cultural tensions, and suggesting that Florus' own interests and legalistic mind-set can be detected in the epistle addressed to him, while Oliensis 1998 argues that Horace uses compliments to the younger Florus for his own self-representation and ‘face’.Footnote 31
The Ars poetica has been much analysed since Brink's magisterial edition, which not only placed the philological interpretation of the poem on a new level but also opened up a number of avenues of literary treatment. Russell 1973 is a sympathetic reading of the poem as both a work of literature and a work of poetics, with some indications of its major influence on post-Renaissance Western culture; Laird 2007 and Reinhardt 2013, noted above, are sure guides to recent scholarly discussion. One key issue has been the poem's structure: most analysts agree that lines 1–294 treat poetry and the poem while lines 295–476 treat the poet, but further divisions are controversial, as can be seen from comparing the detailed structural schemes in, for example, Brink 1963, Williams 1974, Rudd 1989, and Kilpatrick 1990. Interest has also focused on the presentation of the speaker, following the application of persona-theory to Horace's earlier sermones (see above, Chapter II): most have seen at least some autobiographical colour in the figure of the poet who has given up non-sermo poetry and turned to literary didactic about poets and poetry (Ars 306: nil scribens ipse docebo [‘though writing nothing myself, I will give instruction’]), but no self-representation by Horace is straightforward, even if few have followed Frischer's intriguing argument that the Ars presents a ‘mock-didactic parody of a pedantic speaker not to be confused with Horace himself or his usual poetic persona’.Footnote 32 Oliensis 1998 neatly argues that Horace reinforces and advertises his laureate status in the Ars by providing instruction for the young which is as much social and cultural as literary and poetic: ‘Horace is teaching the Piso brothers how to fashion themselves as well as their poems’.Footnote 33
The third-century-ce commentator Porphyrio (on Ars 1) famously reported that Horace drew the critical precepts of the Ars from the Hellenistic Peripatetic critic Neoptolemus of Paros (third century bce), but, apart from a few polemical mentions of Neoptolemus in the literary criticism of Horace's older contemporary the Epicurean poet and philosopher Philodemus, little is known of this shadowy figure.Footnote 34 Philodemus mentions that Neoptolemus treated poet, verse, and poem (or possibly poet, style, and plotFootnote 35 ) as three separate elements; this may be reflected (if somewhat indirectly) in the Ars, which does consider all three elements (on whatever interpretation), if not in this strict order or division.Footnote 36 Neoptolemus also held that the poet should both charm and benefit his listeners, famously encapsulated at Ars 343, omne tulit punctum qui miscuit utile dulci (‘he who has combined the profitable with the pleasant has gained every vote’). Exhaustive analysis of what can be known about Neoptolemus has got no further than these points,Footnote 37 and it seems best to express caution on the details while being clear that the Ars, with its strong interest in imitation and drama, is recognizably in the Aristotelian tradition of the Poetics which Neoptolemus among many others represented.Footnote 38
This prominence of drama in the Ars has often puzzled critics:Footnote 39 why should a lyric poet and non-dramatist spend so much time on plays in a work on poetry, especially in a literary period not really known for its drama (Quintilian's comparative catalogue of Roman literature can only muster Varius' Thyestes and Ovid's Medea for the Augustan years; Quint. 10.1.98). The prominence of the very Greek genre of satyr-play (Ars 220–50) is especially mysterious: Wiseman 1988 has argued for a revival of satyr-play under Augustus, but the evidence is limited and unclear. As critics have often contended, the precepts applied by Horace to drama also make sense for his own non-dramatic poetry, for example his material on generic flexibility.Footnote 40 In both cases the weight of the Aristotelian tradition is likely to provide part of the solution: Horace carries on the primacy assigned to drama by Peripatetic criticism and by his presumed handbook models, in a tralatician move usual in technical works. We might also think that the Ars is happy to return to the classical period of Greek literature and its typical genres, since Augustan Rome could easily be seen by its citizens as a glorious climactic period parallel with fifth-century Athens;Footnote 41 recent research has shown some interest in interpreting the Ars in the light of other Augustan cultural features, such as developments in visual art.Footnote 42 This and other questions will ensure a lively literature on the Ars for many years to come.
