Conviction

Last week, I had a go at the more arty end of writing, the “in a trance” school. This week, I thought I’d alienate the rest of you by commenting on the other end of the scale, the “sausage factory” method.

The sausage method holds that writing is basically like producing sausages on a machine, except with words instead of dubious pork. You find your target market, you figure out what they like, and then you duplicate that, at a fixed rate of words per day. Bingo, instant bestseller.

Or not...

The trouble with this is that it’s nonsense. First, you will inevitably be writing derivative product. This may not matter to you if you’re just in it for the money (in which case you’re a total fool. You’d probably make much better money working in a shop). 

Second, you probably won’t be writing a bestseller. You’ll be writing the thing that people read when they’re tired of the person you’ve ripped off. It will, at best, be second-best and a second choice, and will have to contend with a lot of similar books written for the same reason.

Third, even authors who have a limited range aren’t cynics or hacks. A colleague once said of Lee Childs that “He writes the same book over and over again, but he writes it really well”. Everything I’ve read and heard suggests that Childs takes his work very seriously and is far from “phoning it in”. That’s probably why, even if he does write the same book over and over again, he makes a good job of it. 

 

Conviction

Which takes us onto a wider point. If you look at great “light” novels, the sort of thing dismissed as “genre” or “entertainment”, there’s always something deeper than the need to make a quick buck. You can’t read ‘Salem’s Lot without feeling that Stephen King cares about the decline of small-town America. Rogue Male by Geoffrey Household is full of observations about landscape and social class, as well as being a fast-moving thriller. Tim Willocks’ Green River Rising, a ferociously violent novel about a prison riot, has great sympathy for its heroes, people thrown away by a rotten system.

Perhaps it’s a desire to go slightly beyond the boxes that need to be ticked for the book to work. Perhaps it’s the inclusion of things that the author does care about, and which he is going to make you, the reader, care about even if you didn’t buy the book for that. It's a hard thing to define. The best word I can think to describe it is conviction, and I don’t think a novel can be truly great without it.

"My Characters Surprise Me"

Quite often in fiction writing, I hear someone observe that their characters surprise them. Some writers regularly talk about their characters as if they were independent people, whose actions the writer can't predict.

Personally, I find this slightly annoying: it's rather twee, and gives strength to the false impression that the writer is a sort of conduit through which the story flows, as if the writer's brain isn't controlling the story so much as surrendering to the spirits like a medium. Writing is, I think, a craft, and as such can't be done well by either a mechanical process or subconsciously. The correct method lies somewhere between the two.

(And don't get me started on people who talk endlessly about how awesome/messed up their characters are. First, they're not real, and, second, I don't care.)

However, while I'm never going to like the idea, I can actually see what people mean when they say that their characters surprise them.


Chekov's Gun, In Reverse

Might come in handy later

 

In a novel, it's a pretty basic rule that the characters can't and shouldn't do things that they wouldn't  concievably do. You've probably heard of Chekov's Gun: if the audience are told that there is a gun hanging on the wall, they will expect it to be used. This cuts the other way, too: if the story requires a gun to be produced in Act 3, it should be clear before then that it is possible that this might happen.

So, characters and events shouldn't happen that are completely unexpected, even if that expectation is only one of a number of possible outcomes. In that way, characters are predictable (or at least should operate within a range of predictable outcomes), and shouldn't be surprising the writer by suddenly acquiring abilities and histories that weren't hinted at before.

(I suppose I could imagine a story where a character acted wildly out of character, but I would expect the emphasis of the story to be about discovering why they did that. One example might be a mild-mannered man becoming a terrorist, or a spy being revealed as a double agent.)


A Range of Choices

That said, just because a character behaves in a certain style, that doesn't mean that there's no room for choice. When confronted with a large group of enemies, a veteran soldier might decide to attack, fall back, or just give up: depending on the circumstances, all of these might be decisions that that man might reasonably have made. And sometimes the author decides that one reasonable decision is better to the one that he'd originally planned to use.

Attack!

I think this is what writers mean when they say that their characters surprise them. You reach page 110, when Sergeant Jones is about to die bravely in a hail of bullets, and realise that, actually, it's much more powerful if he decides to surrender and the enemy shoot him anyway.


The Result

Of course, the writer has to be able to deal with the resulting fallout and its effects on the story. The murder of Sergeant Jones means that the enemy now look much more evil to the reader, and the reader is likely to want more payoff as a result. It's important not to lose your grip on the plot or the tone, or to let things done on a whim derail the novel. Because ultimately, no matter what your characters may or may not "decide" to do, it's your book.