Theory of the Lyric 
by Jonathan Culler.
Harvard, 391 pp., £19.95, September 2017, 978 0 674 97970 3
Show More
Show More

Chopping up literary activity into manageable portions of relatively similar material is, like butchery, a job that requires both skill and a measure of brutality. Of all the limbs into which literature has been subdivided by its anatomists, ‘lyric’ is perhaps the most like Grendel’s arm after Beowulf tears it off and hangs it up in Hrothgar’s hall: huge, a bit of a mess, and, in its vastness, terrifying to contemplate. The earliest discussions call this kind of verse ‘melic’ (the Greek melos means ‘song’), and roughly distinguish sung poems from epic and tragedy. Aristotle, who had a strong preference for narrative forms, more or less shrugs off this type of poetry. If he had made a few more observations about lyric then Western thinking on the subject might have been less of a muddle than it was to become. The word ‘lyric’ came to be used by learned critics associated with the library of Alexandria in around the second century BC to describe a broad range of poems written to be sung either by solo voices or a chorus. Lyres were not compulsory, but music seems to have been, and so several kinds of poem which we would now include within our much broader conception of ‘lyric’, such as elegies and epigrams, were put in a different box from odes about drinking or victorious heroes, simply because they were designed to be spoken rather than sung.

The Hellenistic period established a canon of nine ‘lyric’ writers. This included poets who have become household names (Sappho and Pindar), as well as several who have not, such as Ibycus and Simonides – although the latter did come back to life in 1978 as the hero of Mary Renault’s The Praise Singer. Most poems by what the Romans were to call the ‘lyrici’ survive today in fragments, which fuelled the notion that lyric could be swirlingly obscure and passionate. One fragment of Ibycus reads ‘he flies in the alien void’; another is Ezra Pound in embryo: ‘glory … mad … the sting’. Sappho’s deliciously burning fragment 31 (‘no: tongue breaks and thin/fire is racing under the skin’) survived in Longinus’ treatise On the Sublime, which encouraged a belief in the 18th century that lyric in general and odes in particular were a supercharged, enigmatic, sublime kind of verse. The range of dates of the lyrici – from Sappho in the early sixth century to Simonides in the 470s BC – was as wide as their subject matter. ‘Lyrics’ could be paeans to the gods, encomia of kings, victory odes for athletes, general reflections on life, as well as descriptions of drinking or of beautiful girls or boys. But by the first century BC lyric at least had a canon and a set of themes. Roman infants could not, of course, want to be engine drivers when they grew up, but being a lyric poet did become a possible ambition: Horace’s first ode to Maecenas ends by asking his patron (or his readers, or both): ‘But if you rank me among lyric bards [‘lyricis vatibus’]/I shall touch the stars with my sublime head.’

There is no simple historical continuity between Horace’s odes and later metamorphoses of what we call ‘lyric’ in the middle ages, when songs and rhyming Latin verse – either erotic or religious, or, in the case of songs to the Virgin, a dodgy mixture of both – fed the jongleurs as they jongled, and goliardic poets as they goliarded about drink and sex. Petrarch, with deep roots in Provence and in troubadour poetry, fused their eroticism into the tight yet infinitely extensible form of the sonnet sequence: his ‘I’ in love, the distant ‘you’ of the lady, and the yearning between the two, set up the main features of what later ages would call ‘lyric’. It wasn’t until the mid-16th century, though, that a tripartite division of all poetry into dramatic, epic and lyric media – a division roughly sketched out in Plato’s Republic – hardened into a way of carving up all literary activity. The English humanist Roger Ascham could feel pleasingly fashionable when he claimed that his friends at St John’s College, Cambridge, had worked out that all poetry could be divided into the four very large categories of comedy, tragedy, epic and ‘melic’. Lyric became, in effect, the ‘everything else’ of poetic theory, stuff that was metrical and possibly musical in aspiration, but which did not tell a story or dramatise distinct speaking voices.

By the later 18th century this big shapeless bag of everything else had inflated into something which resembled a metaphysical ideal, and floated towards the centre of poetic activity. Hegel in his Aesthetics from the 1820s described the content of lyric as ‘the mind that considers and feels, that instead of proceeding to action, remains alone with itself as inwardness, and that therefore can take as its sole form and final aim the self-expression of the subjective life’. Hegel’s followers, epigones and incomprehenders (his prose does make it easy to be one of those) turned lyric into the mode in which poets represented the operations of consciousness and creativity. Spirit strutted its stuff in verse as the poet poured out his (and it was usually his) feelings.

By the mid-20th century, New Critics had built up a set of depersonalising defence mechanisms against this psychologisation of lyric. Readers were taught to think of ‘lyric’ poems not as the spontaneous overflow of the author’s feelings, but as something akin to a dramatic monologue in which a persona usually called ‘the speaker’ (and the fact that this fictional being was never called ‘the singer’ shows how radically lyric had moved away from music) bewailed his outcast state in a performance of emotion. Students’ knuckles were rapped if they were naive enough to identify the attitudes of ‘the speaker’ with those of the author. ‘Lyric’ in effect came to mean ‘any poem short enough to be set as an exercise in practical criticism’, and included epigrams and elegies and sonnets and poems about what the poet saw out of her window on a Wednesday morning in February. This widening of lyric’s empire to encompass more or less the entire field of poetry led the critic René Wellek to declare that ‘one must abandon attempts to define the general nature of the lyric or the lyrical. Nothing beyond generalities of the tritest kind can result from it.’ He had a point: the word ‘lyric’ was used in such different ways in the second century BC and in the 20th century that joining the dots into a coherent picture requires a broad brush and some pretty bold splashes of paint.

Jonathan Culler’s Theory of the Lyric aims to show that Wellek was wrong, and that it is possible to make transhistorical claims about lyric. Culler is a veteran of the theory wars of the 1970s and 1980s, when he wrote what remain the clearest sympathetic explications of structuralism and post-structuralism. He has an exceptional ability to see the conceptual shape beneath a critical discourse, even if that shape is clouded by jokes or whiffs of bullshit, and to explain it in plain terms. Theory of the Lyric displays those skills. It begins with compressed but beautifully clear histories of both lyric and thinking about lyric; and, like other indispensable studies of this area such as Barbara Herrnstein Smith’s Poetic Closure, it quotes in full a range of particularly well-chosen lyric poems, from Sappho and Goethe to A.A. Milne and A.R. Ammons, in order to establish its claims. The book also displays Culler’s heritage. He positions himself (at times slightly in the spirit of a cold warrior of the theory wars) against two orthodoxies. On his right flank is the New Critical tendency to treat ‘lyric poems’ as quasi-dramatic monologues. On his left is the New Historicist orthodoxy that has more or less taken over the critical academy since the 1990s. This would deny that we can think about ‘lyric’ in the abstract at all, since it would regard instances of lyric as being so embedded in historical particularities that they could not meaningfully be arranged in a single family tree or share any underlying characteristics.

Culler disagrees. He surveys a range of poems that have been thought of as ‘lyric’ and argues that they display a number of common features. Lyrics frequently deploy apostrophe (a direct address to a person or thing, like Lamartine’s ‘Le Lac’ with its call for time to stop, ‘O temps, suspends ton vol’, or Keats’s address to the Grecian urn, or Horace’s address to the fountain of Bandusia). He suggests that lyrics also, rather than being quasi-dramatic representations of a ‘speaker’, often present what he terms ‘triangulated address’, in which an address to a person or thing is designed to be overheard by a reader: ‘What we “hear” is our own ventriloquising of ambiguously directed address, though we may, and in some cases certainly do, construe this as overhearing a distinctive poetic voice.’ So Frank O’Hara addresses a leaf: ‘Leaf! You are so big!/How can you change your/colour, then just fall!’ and John Ashbery talks to a fictional reader who is not there: ‘Why do I tell you these things?/You are not even here!’ These fictional addresses to an object or person in a fictional ‘now’ – lyrics are very often in the present tense – express general truths (leaves fall, readers are not there) and amount to more than a ‘persona’ voicing a fictional state of mind: they draw the reader in to an interpersonal exchange which is sufficiently general to be re-enacted by different participants but sufficiently specific to create a fictional occasion. This gives rise to what Culler sees as a major feature of lyric: what he terms its ‘iterability’. That is, the lyric ‘now’ is not just one occasion but is a representative instance of a possible series. There are many leaves that fall, many ‘yous’ who are in love, many readers who are not there. Lyric, through rhythm and rhyme and other structuring devices, gives to that repeatable moment the weight and iterability of a ritual action – a feature which, Culler would argue, unites the public performances of odes in the ancient world with the private rehearsal of poetic texts in the minds of modern readers.

The chief strength of Culler’s approach is that it helps to explain some of the oddest features of the lyric tradition. It gets inside the strange way that pronouns and interpersonal addresses can function in songs in particular. So when The Beatles sing ‘I want to hold your hand’ probably only a teenager would imagine that they really do mean it’s your hand they want to hold. But one thing lyrics can do is call out to the teenager within by creating social performances which seem to be addressed to now and me: yeah, Paul, you can hold my hand any time, and you can touch me too, if that makes you feel happy inside. The brute fact that it almost always isn’t me who is actually addressed by a lyric pronoun is one of the reasons why lyrics can make teeny-boppers weep with frustration. But they can make adults cry too, as in Keats’s sinister prototext of ‘I want to hold your hand’, in which the reader is set at once inside and outside the range of the poet’s address:

This living hand, now warm and capable
Of earnest grasping, would, if it were cold
And in the icy silence of the tomb,
So haunt thy days and chill thy dreaming nights
That thou would wish thine own heart dry of blood
So in my veins red life might stream again,
And thou be conscience-calmed. See, here it is –
I hold it towards you.

The greatness of this poem (which may have been designed to be part of a play) lies in its apparent ephemerality. It seems to be the work of a moment, which declares itself as belonging to ‘the now when “this living hand” is “warm and capable of earnest grasping” – and the time of reading’, but then pours forth into the future, threatening, drawing life out of its addressee, who is both whoever Keats wrote the poem to (Fanny Brawne as it was once thought, or a character in a drama) and whoever reads it anytime ever, before it tails off into what looks like an unfinished pentameter line, and presumed death. The art of the lyric lies in creating something pitched between the now and the ever, which manages to appear at once universal and an address to me at this moment. This in turn can explain the peculiar pathos a poem can create. In reading a lyric it’s possible to feel absorbed into a universal event while at the same time experiencing the sadness that comes from being excluded from the particularity of the experience. It reaches out across time without ever quite making contact with you or your now.

Culler is also able to explain not just the aesthetic effects but the content of many lyric poems. Why do lyric poets often deal in transient pleasures – the curls of a lover’s hair, the taste of wine, the fragility of rose petals, golden locks to silver turning – but also, often in the same poems or sequences of poems, open up deserts of vast eternity or describe monuments more durable than brass? Culler’s emphasis on the ‘iterability’ of the lyric helps to give coherence to this apparent miscellaneity of content. A lyric poem might represent a moment, but it isa moment so abstractable, through the modes of address it uses and through the generic nature of the situations and occasions it evokes, that it can move between transience and permanence. So ‘dust in the air suspended’ might not just be an evanescent fume, but an event that recurs and reverberates into eternity. Like hymns and psalms and prayers and liturgy, lyrics can make a single moment seem like an every moment: when we sing ‘O hear us when we cry to thee/For those in peril on the sea’ they are in peril on the sea now, they are in peril every time the chorus sings that hymn during the storm in Britten’s Noye’s Fludde, and they were in peril when the words of that hymn were written in 1860 by William Whiting. A poem can let you step into a moment that is always – or as Culler puts it in his not always light-footed prose – lyric ‘is not the description and interpretation of a past event but the iterative and iterable performance of an event in the lyric present’. A lyric poem is itself an ‘event’ not in the sense that the storming of the Bastille was an event; it can enact an interpersonal exchange that occurs at the moment of reading, and can represent actions that could occur again. This is why people talk about poetry taking them ‘out of themselves’, and why reading or reciting a lyric poem can give you a sense of not being here and now, but in a forever zone – and why, for the Molesworths of this world, readers of lyric poems live in a parallel universe along with Fotherington-Thomas and his flowers and birds.

But Culler is no aesthete. In a final chapter he explores how the curious temporal status of lyric can give it a utopian force, as a form that evokes possible other worlds. Poems of praise can create imaginary communities, which extend into the future and construct frameworks of moral value through the offer of immortality. Wordsworth’s Lyrical Ballads encourage their readers to imagine themselves as part of a different world from the one they in fact occupy, which might be pre-capitalist or post-capitalist. The formal properties of lyric, its shape, its rhythms, its ritualised structures, Culler argues, give force to that utopianism: ‘It is when poems establish themselves as memorable, live as poems, that they are most likely to tincture or fracture ideology, to structure our approach to the world, and thus to have a chance of bringing into play their critical edge.’ So we are not just taken out of time by lyric, but can also be made to reimagine our times. Even A.A. Milne’s ‘Disobedience’ (‘James James/Morrison Morrison/Weatherby George Dupree’), though it’s unlikely to change the world very much, may use the allure of its rhythm to make us wonder what James’s mother might actually have been doing when she put on her golden gown and drove to the end of town, and whether or not she should have been doing it.

A theory of anything tends to prompt a search for instances that don’t quite fit, and by the time a theory of anything approaches the scale of a theory of everything exceptions are bound to multiply. Culler’s view of lyric has a flexibility that enables it to stand up pretty well to lyrics that might fly at it from left-field. So the song ‘Yoshimi Battles the Pink Robots’ by The Flaming Lips, for instance, wouldn’t knock it off course. Indeed this lyric seems to fit his model of lyric spookily well: it’s set in a dystopian now, someplace almost here, and has a chorus which includes a transposable me and transposable you, combined with an apostrophic address: ‘Oh Yoshimi/They don’t believe me/But you won’t let those/Robots eat me.’ Culler’s suggestion that lyric no-time can have an influence on the present also casts light on this early 21st-century, drug-soaked version of dystopian lyric, since those robots clearly aren’t at all nice, and Yoshimi (who, we learn, is a black belt in karate) may yet stop them taking over and save the world. It’s lyric, though not entirely as we know it.

So Culler’s view of lyric is a powerful tool for thinking about poems, or at least romantic and post-romantic ‘lyric’ poems. But does it always work? His title modestly or shyly avoids putting either an indefinite or a definite article before the word ‘theory’, and what the book offers is something more like a study of the generative grammar of lyric poetry and of its practice than an all-encompassing model of what lyric has to be or could become. Theories of literary form aren’t like theories of gravity, since they don’t and probably couldn’t have predictive force. Poets make knight’s moves with conventions, taking a rule and twisting it sideways, or transposing and reframing what their predecessors have written. Pound’s famous interlinguistic echo of Pindar’s Second Olympian Ode in ‘Hugh Selwyn Mauberley’ – ‘tin andra, tin hēroa, tina theon,/What god, man, or hero/Shall I place a tin wreath upon?’ – turns the Greek indefinite pronoun tin into English tin, and so reverse-alchemises the crown of praise into base metal. That is one example of how surprising and unruly poetic creativity can be, and illustrates how hard it might be to bind it within a theoretical formulation. But even allowing for this general fact, Culler’s view of lyric does have some blind spots. At the centre of his vision lie French and German Romantic and later poets, who had inherited, via Petrarch and then chiefly through Ronsard and the Pléiade, a model of lyric that foregrounded the drinking, loving, yearning elements in the classical tradition. This line of development effectively sidelined a central aim of much early Greek lyric – to celebrate victories in battle or athletic contests, or the founding of cities. Some of these early panegyrics may have been sophisticated political acts designed to make tyrants change their ways, and so tie in with Culler’s belief that lyrics can act on the world. Many of them, though, weren’t ‘events’ in Culler’s sense, but truly occasional poems which lyric poets were singing – literally – for their suppers, in order to make the men with resources smile. Culler has little to say about this early outer fringe of lyric. He is also a bit reticent about more recent poets who have explicitly broken with the dominant tradition of Western lyric – into conceptual poetry, for instance, or in the work of the L=A=N=G=U=A=G=E poets. How do they fit?

But Culler doesn’t pretend to prescribe the works that people, in the future world in which Yoshimi battles the pink robots, might want to call ‘lyric’. Probably by then zeros and ones will have eased out the ‘ohs’ and ‘ahs’ of romantic lyric forms of address, and people, if there still are people, will be looking for things in verse that we can’t yet imagine. Lyric poetry is a practice that depends on grasping both prior instances and the implied general rules that underlie those earlier instances, and then shaking the whole lot up to make something akin to but different from those earlier instances. Along the way, things get misunderstood, new language games are brought into the mix, hitherto neglected poems suddenly come to seem important, and the whole bangshoot might get transformed. Culler’s subject is that tradition as it has evolved in the West, which is why forms of lyric that turned out to be historical dead ends or that consciously shatter the tradition aren’t part of his concern. But, who knows, robots and tin men in the future may well hum songs of praise about the people they ate and the cities they destroyed, and call them lyrics.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 39 No. 19 · 5 October 2017

Colin Burrow writes that ‘lyrics are very often in the present tense,’ and that they use a ‘fictional “now"’ to express general truths (LRB, 7 September). In English, we very rarely use the present tense to talk about ‘now’. Speakers often use the present tense to relate experiences in the past, and we much prefer the present tense when talking about the future. When we say, ‘My dog eats chocolate,’ we invariably do not mean it is eating chocolate now – it isn’t, look – but that it has done so in the past and will, we expect, continue to do so in the future. Cows eat grass, Peter makes mistakes in his maths, the sun also shines.

In poetry workshops these days, one is politely discouraged from using the ‘continuous present’: don’t say ‘the leaves are floating,’ but ‘leaves float.’ The simple present is grittier, one is told, more direct, more condensed. What’s really meant is that it gives the idea of a ‘general truth’. It expands, exactly as Burrow explains, a simple idea into a gnomic piece of wisdom.

Wordsworth’s ‘Daffodils’ uses tense precisely. He tells his anecdote in the past tense, but gives his reflection in the simple present. He did this; this is the effect of that. By linking the past to a general truth in this way, Wordsworth expresses his idea with a certain humility. Modern ‘lyric’ poetry rarely follows this model, preferring to endow the simplest experience, as Burrow says, with overtones of eternity.

Philip Rush
Stroud, Gloucestershire

Vol. 39 No. 20 · 19 October 2017

Philip Rush points out that we don’t use the present tense much to talk about ‘now’ (Letters, 5 October). He might have added that even when we do talk about ‘now’ we tend to use the continuous present. This wasn’t always so, but today Puck would probably say: ‘I’m going, I’m going; look how I’m going,/Swifter than being on a mighty Boeing.’ (In fact, he’d probably be like: ‘I’m going, I’m going, keep your hair on.’)

Beverley Rowe
London NW1

Vol. 39 No. 21 · 2 November 2017

It is not correct, as Philip Rush puts it, that English speakers ‘much prefer the present tense when talking about the future’ (Letters, 5 October). You might as well say that English speakers prefer to use the singular for the plural in nouns such as ‘deer’. The construction ‘My dog eats chocolate’ is not an example of either the present or the future tense, but of the consuetudinal tense, referring to that which is constant, habitual or naturally the case. In Germanic languages the consuetudinal tense takes the same form as the present tense, while in Welsh it takes the same form as the future tense. It is interesting that English has reverted to using the same form as the present tense, whereas Old English or Anglo-Saxon (spoken in England until about 1150) used the same form as the future tense. The influence of Old Welsh on Old English?

Margaret Faull
Wakefield, West Yorkshire

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences