Disagreeing with Kapka Kassabova (and thanks to Ian Rankin)
Are you a course junkie?

Keeping up with the Jameses

Someone reading my post In Praise of the Long Sentence recently took issue with one of my examples, taken from un(der)educated, 15-year-old Anna's narrative in The Mathematics of Love:

They were tiny of course like all the other negs I'd looked at, but different because I was looking at them in one curling strip and all still wet: clear lavender-coloured shadows and dark skies, trees and pillars and windows and faces caught click after click, coiling and springing down the film one after the other so that all the distance and time between them was pressed into plain, pale bands of almost nothing.

Would an uneducated fifteen year old really talk like this, they asked; would she be this articulate, and in a sentence this long? I had gone very carefully about the business of finding a voice for Anna which was convincing (loose grammar, no metaphors but only similes, unsophisticated ways of saying things) but could also be as vivid and evocative of both things and ideas, as I needed it to be. But this question got me thinking, because of course Anna doesn't talk like this: this is not how she would say the same thought aloud. So how can her narrative voice - as opposed to her dialogue voice - be like this, if this isn't something she would say?

It's clear enough that if a narrative is in third person, then there's some kind of implied, external narrator, and that narrator by definition has a voice - a particular way of saying things - which may or may not be the writer's natural way of saying things. How much and when that voice takes on the colour of one or more characters' voices is the whole game with free indirect style, and that flexibility is one of the great advantages of a narrative in third person. But I would argue that even in a narrative in first person, it can be very useful to think of character-narrator James, say, as a subtly different entity from the James who inhabits the same novel-world as Jilly and Jonathan. The obvious example would be where James, now older, re-tells what happened in the past: in a kind of free-indirect-first-person-style, to a greater or lesser degree (and probably variably) James bring his older ways of describing and perceiving things, to the business of evoking that younger James. But even when a narrative isn't explicitly structured like that, in telling his story James is acting almost like an external narrator, putting words to things which the James in the drama wouldn't say out loud, or not like that. Narrative is not the same as dialogue, in other words.

What we're doing with a narrative voice, it seems to me, is not unlike what a painter does, as opposed to a photographer. Failing Photoshop, a photographer can only reproduce what is actually before the camera, although they may set that up very carefully. That, it seems to me, is a bit like writing any naturalistic dialogue: the goal is to keep us in the forgetting-to-disbelieve world of the novel, in which we take these things as actually being spoken. On the other hand, a portrait painter starts from the features before her, but uses paint and technique to say something less straightforwardly reproductive, and more expressive; the painting might be less useful for meeting the right person off an aeroplane, but we don't take the swirls and bumps of paint or the non-naturalistic colours literally; their job is telling us other things (or getting us to sense them) about who this person is.

And yet we must also recognise, within that expressive paint, something essential about the face and looks. Similarly, we don't take the narrative voice as words actually spoken aloud, and it can therefore range wider than strict verisimilitude would permit, but it must also have something essential in common with the dialogue voice, or we won't be convinced that the character and the narrator are actually the same person. To keep your reader forgetting to disbelieve that this person is telling us what happened, you need to make sure that both speech and narrative spring from the same real-seeming consciousness.