Poetry and precision

In poetry reviews and blurbs, the terms "precise" and "precision" are commonly seen though their use is often vague. They seem to be terms of praise hinting at 2 main concepts -

• Accuracy - The ability to judge this rather depends on shared understandings and mimetic intent. Is Pound's "In a Station of the Metro" accurate? When Gerard Woodward writes "a toilet cistern refills like an old lady pouring tea" is he accurate? If you see a fuzzy painting of a landscape, it could be the result of the artist's lack of attention to detail or an accurate depiction of a foggy day. How is one supposed to know?
• Economy - "Precise" often means "small" or "clipped". When an artist captures a face with a few brushstrokes we admire the precision. Economy in poetry tends to involve word-count rather than line-count. 10 words spread over 5 lines is more likely to be considered "precise" than a 10-word sentence. Neither is the time taken by the reader taken into account. If twice as many words mean that a reader understands in half the time, there's a case for saying that the longer version is more efficient, more economic, but it wouldn't be described as more precise.

"Precision" to me can apply to big or small things. There needs to be a small margin of error - if the thing were different even in a small way, the effect would be significant. This concept of "precise" isn't the one that's usually applied to poetry, especially when line-breaks are praised for their precision. I think the economy of the image, more than its exactness, is what provokes the use of "precision", which is why haiku, short Imagist poems and short-lined short stanzas are most likely to attracted accolades for precision. A single, well-aimed (preferably striking) image can be more effective than carpet bombing, and an image can be made to look more significant than it is by being isolated.

The easiest situations to claim that words are precise are where other media have trouble competing at all, or where words' power of reference and allusion (rather than expression) can be exploited. The leverage of a phrase like "remember that night under the bridge in York?" can be large if there are shared understandings or if preparations have been made earlier in the text. A single word or name can be the key to a store-room of memories.

But if one wishes to be precise, why restrict oneself to using words? Words often can't compete with other media. In his novel, "A Heartbreaking Work of Staggering Genius", Dave Eggers includes a floor-plan - a more precise way of describing the house-layout than words would be. And a photo of a delicately coloured rose is going to convey appearance faster and more accurately than words.

The mouth of Franz Hals' "Laughing Cavalier" is a smear of red highlighted with what looks like toothpaste. Close up you'd be hard pressed to say what the marks represented. By 1851 that style was out of date. Late one summer's evening, Millais rowed across a river to sketch the flowers on the far bank by lamplight. You'll see those impossibly detailed flowers in the Tate's "Ophelia". In contrast there's Picasso's drawings - bare lines on a white expanse - and Schiele's work where areas of untouched canvas compete with highly wrought flesh. Later still came hyper-realism, where painting aspires to photography.

When a poet writes "he rowed across the river to paint flowers", readers are trusted to fill in the details if they wish. It doesn't matter if they erroneously think that daffodils proliferate on the riverbanks. The poet's "flower" is that of a toddler's first doodle: a generic icon. If the writer says that the flowers are "yellow", we wonder why that feature is emphasised. Writers cannot mention without pointing, whether it's at objects or their secondary properties. Because of that, poets have to be as careful as pre-budget chancellors about what they say. Every word matters, but they don't tell the whole story. In contrast, painters cannot paint mere "flowers" - they have to be yellow, drooping, or windblown. Because there are always so many details, none can be singled out.

The world's becoming more graphic, less poetic - we are offered a choice of details to concentrate on rather than being trusted to fill them in. Any lack is a vacuum that must be filled: we have to know the history of each killer, interview those they knew at school, know in which beauty spot the body was buried. If we mistrust what we see our response is to zoom in. There are always more details to find, more trees to obscure our view of the woods. Yet we are scared to get away from it all. Walking in the countryside, we take along our binoculars, our mobile phones and Sunday papers. Even if we notice hosts of daffodils we would no longer describe them as golden - in truth they aren't. Yet facts are often no more than props for poets to feel their way into a poem, scaffolding. Like an actor's false nose, they are as much for the performer as the listener. Like experimental findings, they are useful only for what we can derive from them. We are becoming obsessed by the fine print and the appendices, forgetting the abstracts. Too many writers play safe by giving us everything on the principle that more is better, betraying a lack of self-confidence, an inability to select. The law of diminishing returns applies, further details becoming exhausting rather than exhaustive, obscuring rather than illuminating the original, squeezing the reader out of the text. Picasso was right when he said that it takes a master to know when to stop.

Art ranges from Pure Abstraction to hyper-realism. Poetry too has abstract forms (Schwitters' sound poetry for instance) but its range (excepting perhaps dialogue) doesn't extend to the reproduction of the real world. Poetry has to accept that it can present appearance little better than it can smells, that because the natural world has to be translated into words, all aspects of reality are equally available, equally distant. Poetry needs to combine the untouched abstraction of "flower" and "summer" with the selective power of adjectives, the pinpoint precision of quotation, and especially of proper nouns. The blanket-bombing of Millais is absent. Instead, details are pruned back to let the spaces speak, and lines define not just area but volume. Under magnification the phrases and words may look mundane or even careless but the pieces aren't meant to be observed in isolation. Each word is modified by its context. Meaning is distributed, oblivious to word boundaries and even the boundaries of the work. When a modern poet writes "golden daffodils" the extra meaning isn't discovered by zooming in on the individual words, but by panning out to take in Wordsworth's poem and our response to the Romantics. The measure of a poem's precision is not the amount of detail it contains, but how well it targets the factories of knowledge in the hinterlands of the reader's mind, where the details are best left.

When I began researching to complete this article I discovered (not for the first time) that Jim Murdoch had already dealt with the topic in more depth than I can manage. He too points out that people often meant "concise" when they use "precise". The term was perhaps imported into literature when there was a trend for borrowing from technology - Futurism, Bauhaus, etc - but without the extremism of Minimalism. He also quotes Marianne Moore's "What is more precise than precision? Illusion". Magritte's train coming out of the fire-place needn't be any more realistically depicted than it is. Too much detail would distract. I think that even fuzziness has its uses -

• One way to make something into art is to remove its purpose. A potter makes a cup so that someone may drink from it. Take away that use and viewers will look for other types of meaning. An instruction manual or a recipe can have its purpose and detail blurred so that only the rhetoric remains.
• A blurred image can be more potent than a "precise" one because the details may be irrelevant to the effect, and the audience can fill in details themselves if they need to, personalised. A dark object at night is more scary if you can't see that it's a bush rather than a hunched figure. A photograph of Christopher Lee playing Dracula may be more effective if blurred so that you can't identify the actor, only the identifying characteristics of Dracula.

And the content may not be the most significant aspect of an intricate passage. It's sometimes useful to give a sense of precision in order to vary the texture - sweeping from grand abstracts to miniaturist detail can be effective, like jerking a microscope slide into focus.