I’ve covered this subject before, but it was brought home to me again while I was working on chapter 21 of Heart of the Fey. A crucial geological detail, essential to the plot, helps explain why the Heart had been lost for so long, and gets my hero on the right track to at last find it. If there were no waterfall, no vanished ford, no barrier between north and south, then there would be no story.

My problem was that, while I had worked it all out intellectually — times, places, history, why they’re there, where to go next — I discovered when I was revising it for first and second read-alouds that it made no sense. It didn’t work physically, geologically, or any way at all. Because I had only intellectualized it, I hadn’t actually visualized it. And that became clear while I was trying to write what my characters saw, because I couldn’t see it, and therefore it didn’t make any sense. I had to stop, and rework the map, and rework the history, and rework other stuff, until I could see it as if it were a video. Now it all makes sense, but it took me over two days to fix one paragraph, and to deal with the consequences of those changes.

Why go to all the trouble?

Because glitches like that can ruin a book, no matter how good the rest of it is. Some time ago I was reading a non-fiction book, which I cannot remember now, but I was interested in the development of the author’s argument, and I got to the last three pages, where he was supposed to at last have it all come together and prove his point. But it was as if I were reading notes instead of completed text. Nothing came together and all, the feeble point he did make sort of contradicted what he had promised, and he lost me completely.

I was terribly disappointed because, whatever it was he was trying to say, the ideas had been good, and worth my time. But the let-down, because of a terrible glitch in logic or reason or follow-through or whatever, was so bad that I put the book away and immediately forgot it. Except my frustration and disappointment. I can’t find the book again.

My wife was reading a novel by an author she enjoyes, but somewhere in there Lady Rutherford was called Lady Blair, or something like that, and the sister of a nobleman wanted her son to inherit her brother’s title, which was not possible — nephews can not inherit — and Diane just dropped out of the story. She finished it, but she wasn’t in it any more. She told me about that while I was working on my glitch, but she can’t remember the author or the story.

And If I didn’t fix this one scene, readers would be ejected from my story, might not finish it, and might even forget what the story was about, except for the glitch. And might decide not to read anything more of mine.

I have a rule. If I write something in a story, which I discover later takes me where I didn’t expect to go, I will work with what I have and keep going. So when I found that the waterfall into the river couldn’t really be that way, I felt compelled to work with it anyway, and to somehow make the story fit. And I couldn’t. Geologically, it made no sense.

But I have another rule, given me by David Gerrold, and that is, whatever you write must serve the story. The story can’t serve the false image, the bad idea, the gross error in creation or research. Everything must serve the story, not the other way around. So despite the first rule, I knew that I had cast aside what I had done wrong before I could go on. Or my readers would discover how wrong it was, and put down the book. And might remember how wrong it was when they were looking at something else by me.

The details of the error are not important here, only in the context of Heart of the Fey, where they eventually affect and color almost the whole story, sometimes retrospectively. The details of the correction are not important here, only that I made myself take the time, almost two whole days, to fix a paragraph, and deal with the consequences of that fix. What is important is that I had forgotten that, if I can’t see it, I can’t write it, especially when it comes to physical description. What I had written was only what my intellect had created, and no matter how interesting that was, it was wrong.

Which is why I draw maps. I draw floor plans. I create timelines. I chart people sitting at a table so that they aren’t talking through each other.

And when I corrected the falls, the geology, the history, all of which fit the story so far, I made the waterfalls more spectacular than they had been before. I made the mystery, of where the Heart had been lost, more logical when it was finally traced and found. It made my characters’ reasoning actually work.

So then. Make use of what I’ve got, even if it takes me somewhere unexpected. But if it takes me into a black dead end, work it until it serves the story, whatever that takes.

I hope to finish Heart of the Fey some time this spring or summer. We’ll see.