Horace is my favourite of the classical Latin poets. In fact, the first Latin poem I read as a teenager was one of his — Odes 3.13, ‘o fons Bandusiae’, is included in the Oxford Latin Course (or at least it used to be). You don’t forget your first encounter with something this good and I remember exactly where I was (in a beach-hut in Suffolk) as I picked my way through the intricate & beautiful Latin of this little poem.1
Horace is also important to me because his odes, in particular, have for centuries lain close to the heart of poetic practice in multiple European languages, and especially to the challenge of translation. It is a peculiar feature of Horace that he has been extraordinarily widely translated and imitated – up to around 50 years ago it was hard to find an Anglophone poet who hadn’t had a go at him – but that the great majority of the poems that ‘imitate’ Horace are not actually much like Horace at all, I mean as a reading experience. This is even true of most translations: it’s as if what we think a Horatian ode should be like gets in the way of what it actually is like, even for translators; or perhaps just that there are so many different and in some sense competing elements to the average Horatian ode, that poets find themselves forced, by their own tradition as well as preferences, to choose between them.
This has something to do with the commonplace that Horatian lyric is (even more than most poetry) ‘untranslatable’. When critics talk about Horace’s ‘untranslatability’ they usually point to particular features of his style, such as the ‘mosaic’ effect of his word order. Latin is a highly inflected language — meaning that the endings of words change in order to indicate various grammatical functions which in English we often indicate by the order of words and with various ‘little’ words, like to, by or of. As a result, word order is more flexible in Latin in general, and Horace pushes that to an extreme in his lyric verse.
It’s hard to convey in English what this ‘mosaic’ effect is like to read, though sometimes the failures of translators give us a clearer sense than their successes. John Milton’s well-known version of Horace Odes 1.5 (the ‘Pyrrha’ ode, so-called because it’s addressed to a woman with that name) contains a notoriously obscure line:
Who now enjoys thee credulous, all gold,
Most English readers, I think, would tend to assume at least initially that both ‘credulous’ and ‘all gold’ describe the addressee of this line, the thee of the poem — that is, Pyrrha herself. In fact, as is clear in the Latin by the different case-endings, thee (Pyrrha) is all gold but it is the Who of the line, a man or young boy, who is credulous in his current enjoyment of her:
qui nunc te fruitur credulus aurea
Here with its characteristic economy the Latin tells us that the who of the line (qui) is masculine and singular, that he is the one doing the enjoying (fruitur) and that he has been overly trusting (credulus) in doing so; while the you to whom the poet speaks is both singular and (aurea) unmistakably female. The juxtaposition of credulus and aurea, meanwhile, activates more general associations: what happens if we trust too much in wealth or appearance.
But Horatian resistance to translation is not just about word order. The peculiar tone and what we might call the tonal movement or tonal juxtapositions of the average Horatian ode are a particular challenge. The Pyrrha ode adopts what appears at first an objective tone, as if the speaker of the poem were not directly involved in the love affair he describes and the young man’s imminent disappointment. But the poem ends quite differently:
miseri, quibusintemptata nites. me tabula sacervotiva paries indicat uvidasuspendisse potentivestimenta maris deo.
Translated by Milton as:
Hapless theyTo whom thou untried seem’st fair. Me, in my vow’dPicture, the sacred wall declares to have hungMy dank and dropping weedsTo the stern god of sea.
Or, as Basil Bunting puts it in my favourite, albeit elliptic, version, spoken in the voice of a mackerel fisherman:
Remember us to the teashop girls.
Say we have seen no better legs than theirs,
we have the sea to stare at,
its treason, copiousness, tedium.
(This is the end of Bunting, Odes 1.22.)
Horace is stranger, flintier, more evasive and austere, less likable and more religious than we think. (Although it is fashionable to ignore them, he wrote a considerable number of hymns or odes with hymnic elements, a feature of his work that was very important to readers in early modernity.) His poems never end quite where we expect them to, and almost never where they started. This adroit sliding is captured very well by Donald Davie who wrote a whole poem about it. ‘Wombwell on Strike’ begins:
Horace of course is notA temporizer, butHis sudden and smooth transitions(as, into a railway tunnel,then out, to a different landscape)it must be admitted elide,and necessarily, whathappens up there on the hillor hill-ridge that the tunnelof syntax so featly slides under.
The omissions Davie was thinking of were primarily political, as the rest of this fine poem makes clear. The final image is unexpected in a genuinely Horatian way, while also claiming to return to the poem the reality of political enforcement which Davie accuses Horace of eliding:
Yours was solid advice,Horace, and centuries haveendorsed it; but over this tunnellarge policemen grapplethe large men my sons have become.
Davie was a fine poet and an astonishingly good critic — I’m not sure I can think of anyone else whose writing about poetry I’ve found so consistently generous and useful. (And there’s a timely new edition of his Selected Poems from Carcanet, edited by Sinéad Morrissey.) ‘Wombwell on Strike’ is itself an extremely effective demonstration of and mnemonic for a particular feature of Horatian style, as well as a critique of it. Davie has Horace articulate in this poem principles he takes to be central to his lyric, though never directly expressed: ‘Press on . . . Leave the unlikely meaning / to eddy, or you are in trouble.’ But as a result the poem is in most respects nothing like an Horatian ode at all.2 Even that genuinely surprising ending works in part by pulling out the emotional stops — the sudden appearance of the poet’s own adult sons — in a way we never find in Horace. The miracle of Horatian lyric, by contrast, is how often the poet concludes a poem with the most ordinary of words, made both beautiful and painfully moving by the whole shape and momentum of the poem they bring to an end. ‘O fons Bandusiae’, for instance, ends not with the young sacrificial goat, just growing horns he is destined never to use, but with the spring and the poet. The very last word is an ordinary pronoun, tuae (‘your’), referring to the spring’s own waters:
fies nobilium tu quoque fontiumme dicente cavis inpositam ilicemsaxis, unde loquaceslymphae desiliunt tuae.
I like Eavan Boland’s version of this poem a lot, but it can’t resist doing what Horace does not, and ending not with the spring, but with the poet himself, and the glory he confers upon it:
With every fountain, every springOf legend, I will set you downIn praise and immortal spate:These waters which drop gossipingTo ground, this wet surrounding stoneAnd this green oak I celebrate.
That’s not to say Boland is wrong: her translation brings out and makes clear for the reader much of what is delicately and even ironically implicit in Horace — the sort of thing a good commentary fills in. Horace’s implicit promise to immortalise a random, unknown spring, to make it as familiar as Arethusa or Hippocrene, the fountain of the Muses on Mount Helicon, is both ridiculous and, ultimately, of course, fulfilled. But the shape of Horace’s stanza is quite different: it begins with that promise, stated simply (fies nobilium tu quoque fontium / me dicente; ‘you too will become [one] of the noble springs / because I am describing . . .’) but then moves away from it, and the final movement of the poem is down and away — lymphae desiliunt tuae, ‘your waters leap down’. The verb desiliunt catches immediately the image of a spring with a drop from rock to rock; but also takes us back to the doomed goat. The waters are loquaces, literally ‘talking’ — as we might speak of a ‘babbling brook’. Boland’s version brings the vignette to a standstill, a stationary snapshot of each element: she grounds the end of the poem (in Horace, the waters just leap down, not ‘to ground’) and lands finally on the unmoving elements — the stone and the oak.
Boland’s translation fills in very effectively a lot of what you could get from a decent commentary: obviously, this is a poem about the poet’s own power to immortalise, to set down, however much the evasive shape and music of the Latin cuts across that. Davie’s ‘Wombwell on Strike’, on the other hand, will undoubtedly make you better able to appreciate Horace in Latin, if you can read Latin or are learning it. But if you don’t read Latin and you want a glimpse of what Horace himself (not the commentaries on him) is like to read in the original, of the least translatable elements of his music and the compression, wit, strangeness and beauty of its juxtapositions and transitions, you should read Bunting, for my money and by some margin the most genuinely Horatian lyric voice to have been achieved in English.3 I will come back to Bunting — and certainly also to Horace — another time.
If you want to read the poem and have any Latin, there’s a text with a usefully literal translation here. As always with Horace, there are many many English verse translations, ranging from the literal to the impressionistic: for the best chance of a glimpse at the original, I recommend reading several. Conington’s is online. Carne-Ross’s anthology Horace in English is excellent if you can get hold of a copy second-hand. You can also hear the Latin of Odes 3.13 spoken aloud here.
Horace’s most significant statement about lyric style, Odes 4.2, is framed as advice to a young poet not to imitate Pindar (while actually doing so).
Bunting’s collected poems include a series of ‘Overdrafts’, his neat coinage for versions or translations, including several from Horace. Several of these are very good to teach with, but the two books of published ‘Odes’ are more impressive and consistently more genuinely Horatian. Oddly, there has been very little good published criticism on Bunting and Horace.