Telling a story requires us to juggle numerous things at once. The best example may be writing conversations between two or more characters.
Dialogue scenes make up a good bit of most books (and almost the entirety of most stage plays). A dialogue scene has to perform some function in the story, or it shouldn’t be there at all. But it also has to be realistic—it has to sound like the way people talk. Not exactly the way people talk, of course. In fiction we strip out the hemming and hawing, the repetitions, the false starts, that one sees in a sociologist’s literal narrative of actual conversation. But it has to flow the way real conversations do.
How can we achieve both things together—story function and plausibility?
The purpose of a discussion scene in the story may include things like advancing the plot, displaying and developing the characters, and providing necessary background information. If the scene belongs there at all, we probably have an idea what at least some of those purposes are by the time we’re ready to write the scene.
I find it useful to note down in advance what points I want to cover in a conversation. Let’s say this scene has to get A to persuade B to do something. How will that happen? What are some of the things A can say to get B moving—logical reasons, spurious reasons, deceptions, appeals to emotion? What are some of the things B might say in response?
We can brainstorm such possibilities and note them down in bullet points. If there are side issues that can be addressed in the course of the discussion—showing what A thinks of B, or what B is afraid of—put those down too. The characters may also be doing something at the same time—not just sitting around a conference table, but eating a meal, hacking their way through a forest, building a Lego model of the Empire State Building. If any of that is relevant, or useful, or entertaining, note it down. We’re compiling points to cover for plot purposes, as well as good bits that we may want to get purely for entertainment.
(Of course the bullet-point method is just one way, not the only way, of starting to plan a conversation—but I find it a useful technique for getting all the raw material together in one place.)
So we’ve got a list of points that we want to get into the dialogue somehow. But how does the conversation flow? How do we segue from one topic to another?
Up to now, we’ve been planning out the scene in much the same way we would outline an article or presentation in discursive writing—nonfiction. But here’s where the approach needs to be different, because making the scene believable means tracking the way a conversation actually meanders along.
An article or advocacy piece proceeds logically, building up in organized fashion to a conclusion. But people don’t generally talk that way even when they’re debating. Conversation proceeds not by the logic of argument, but through something more like free association. Something A says makes B think of an independent gripe, and the smooth progression of A’s argument collapses. Emotional connections drive the developing dialogue as much as logical connections do.
When we sort the bullet points into order, then, in preparation for expanding them into actual narrative, we want to do it based on the way conversations develop in our own experience, as adapted for the specific characters and situation in the story. Where might these characters start? What would come up next? How can we put the conversational moves in order so they all get covered—like drawing a curve through a certain set of points?
In shaping this process, it’s great to be the author, controlling what everyone says. In real life, no one participant controls the whole exchange (much as some of us might like to). The author, on the other hand, can exercise backward causality or teleology. We can have the earlier lines shape the conversation so it arrives at a particularly good line or key point—as long as each transition is a plausible reaction by the characters to what has been said before.
Once we start writing the conversation, converting the bullet points into actual narrative, things may—and will—change. When we get into the flow of the dialogue, we often find that a transition that initially seemed natural sounds forced, and we have to go about it another way. Or we discover that the way an exchange turns out, there’s a quicker way to get where we want to go. We see an opportunity for a really great line—can we work that in, or will it redirect the discussion too much? As in most things, we need to remain free to improvise as the scene develops.
I mentioned the chance to get in a “really great line.” This reminds us that we also need to be thinking about whether the dialogue itself is entertaining—enjoyable to read.
This may be partly a personal bias: I love witty dialogue. But the point has broader importance. I’ve plowed through a lot of writing where a conversation advances the plot, but it’s tedious, dull, uninteresting to read. If we as reader are just suffering through the dialogue so it will lead us functionally to the next scene, something is missing. After all, the reader’s primary purpose is not just to get to the end of the book, but to enjoy the ride.
To my mind, then, we must always keep a triple view in mind when writing a conversation: (a) What would move the story along, serve its purposes? (b) What would come naturally to the characters in this situation—what would they be likely to say next? (c) What would entertain the reader? Good dialogue must satisfy all three criteria.
To put it another way, we should ask ourselves what a conversation is doing in the story both from the author’s point of view (the story purpose), and also from the characters’ point of view (why are they saying these things at these moments, what is the natural drift of the dialogue).
Of course the above represents the observations of a relative novice. For some really cogent discussion of dialogue from a professional writer, I recommend a series of Patricia Wrede’s blog posts, starting here (12/2/2016) and continuing across the five following entries, ending here (1/6/2016).
(For an amusing image of how not to build a conversation, I refer you to a classic Marvel spoof back in 1967 . . .)