Advance planning

There are a lot of bits of advice floating around for would-be, wannabe, and newbie fiction writers who are having problems getting started. There are currently a plethora of how-to-write books advocating serious advance planning, ranging from detailed outlining systems, to starting with different sets of character archetypes, to complex setting and worldbuilding development, to one that mandates how many viewpoint characters each book should have given its intended word count, which scene each new viewpoint character should be introduced in, and how often the author should rotate from one POV to another.

They all have one thing in common, and they all fall apart at exactly the same place. The thing they have in common is that all these systems are a way of making assorted writing decisions in advance of actually writing anything. The theory seems to be that what gets new writers (or any writers) overwhelmed and knotted up is the plethora of decisions that have to be made: who the viewpoint character is, where the story takes place, what all the relationships between various characters is, who can/should be in the story and why, what sort of plot structure to use, etc. ad infinitum. Logically, then, if the writer is encouraged to make all these decisions in advance, it ought to then be easy to sail into the story and out the other side. Some of these systems appear to imply that once you’ve worked through their method, the only thing slowing down your production of your novel will be your typing speed.

There are two major problems with this theory. (Well, three, but that gets into where and why they all fall apart, which I will come to in a minute.) The first and largest problem is that not everyone works the same way, and, in fact, the same author won’t necessarily work exactly the same way from book to book. There are plenty of authors for whom any advance planning at all is the kiss of death to their story. Their process is “sit down in front of a blank page and surprise myself.” I’ve done that twice, and for me it feels like working without a net, but Talking to Dragons and Sorcery and Cecelia are two of my most popular books, so it obviously worked better than I was afraid it would.

There are also authors who need more room to maneuver than these systems allow for. They need to be able to get to the scene in the middle of the novel where the heroine needs to find out some critical bit of information and make up something – trained rats, a reverse-hacking computer program, a convenient Oracle – that will let her do it, and then they backfill some mentions of it earlier in the book so it doesn’t look as if it was made up on the spot. This particular breed of author can’t seem to do this if they’ve tied down all of the background, the technology, the backstory, the rules of magic, and the relationships between all the characters in advance.

Which brings us to the second major problem with the theory, which is that making decisions in advance simply in order to have decided something means making arbitrary decisions. “Decide what each and every character wants,” the system says, so the author goes down the list assigning a want to each character just to have something on the list. They then start developing a plot based on Character Q wanting a house in London, all the time knowing somewhere in their heart of hearts that “wants a house in London” was pulled out of their left ear.

Arbitrary decisions have no justification behind them, and they are therefore easy to change. This is good in one sense, because it means that the author can let go of them when she is suddenly hit by inspiration and realizes that Character Q really wants to live up to his father’s image, and the house in London has nothing to do with anything. It’s bad in another sense, because it means that every bit of backstory, plot, and character interaction in the plan that depended on Q wanting a house in London is no longer valid. This can unravel major chunks of plot, especially since that particular sort of epiphany usually strikes in the mid-book when the writer has really been living with and writing the character for a while and has gotten to know them.

And that is the ultimate problem with all these systems: every last one of them reaches, sooner or later, the point where they have to say “Now, sit down and write the first scene.” And not one of them is any help at all with actually writing the darned thing. It’s still a matter of butt in chair, fingers on pen/keyboard. The writer may know exactly what happens, but an outline that says “Matters reach a crisis when Uncle Earn tries to have five-year-old Eff arrested for setting a curse on his house” is vastly different from actually writing the seven-page scene in which Eff and her parents face off against Uncle Earn and a random policeman.

No matter how many things one has decided in advance, each chapter, each scene, each paragraph still has to be written one line at a time. No matter how much the writer knows about magic or the personalities of the people involved, there are still word-by-word decisions to be made: who will speak first, Eff or Uncle Earn or her parents or the policeman? In what tone? Exactly what does Uncle Earn say; “Officer, arrest that child!” in a commanding tone, or “Eff, why did you curse my house?” in a sad one, or “She’s the culprit!” in an accusing one? Exactly how would he phrase it? Who answers him, in what tone? Does anyone interrupt Uncle Earn or the next speaker? Since Eff is the viewpoint, where is her attention and what does she notice first about each of the people who is there? How does she describe what she sees and feels?

Knowing how the Rothmer sitting room is furnished does not help with any of these decisions. Knowing what the characters are like may help somewhat, but it’s still not going to make any of them easy, especially since most writers I know don’t think in terms of making thousands of tiny decisions, they think in terms of “OK, how do I start this scene? What happens?” When the fingers hit the keyboard, it’s usually a matter of feel and intuition, not word-by-word micromanagement.

A more detailed summary of the action (something like “Eff is summoned to the sitting room, where she finds her Uncle Earn and her parents waiting with a policeman. Uncle Earn accuses her of cursing his house and demands that the policeman arrest her; after a short argument between Eff’s father and Earn, the policeman announces that Eff is too young to be prosecuted under the law and leaves. Uncle Earn persists in his accusation until Eff’s mother steps in, informing him that she has had enough of the way he and his family treat Eff and her brother. Eff’s father agrees, and Uncle Earn departs in disarray”) would have been a great help to me if it were the sort of thing I can do in advance; unfortunately, I can’t ever seem to get down to that level of detail until after I have actually written the scene. And while it lays out the main action for the scene, it still doesn’t do anything to help with the actual phrasing of the dialog, or with the balance between dialog and narrative, or with the ebb and flow of the physical action and the emotional reactions of each character.

Planning is different from actually writing the scene. It can help, if you are the sort of writer whom planning helps (I find it very useful, to a point), but you still have to write the scene. And I don’t think there can be a system to walk writers through the actual writing part, because if there were, somebody would write a program that could do it and we’d all be out of a job.