Writing Excuses 14.35: What You Leave Out
Key Points: Worldbuilding an iceberg? Just build the tip of the iceberg, and make readers think the rest of it is there, too. Build what’s needed for verisimilitude. Figure out where your scenes are set, then figure out what that looks like and how it works. What are you going to be using the most? What will my characters be directly interacting with? Give the reader information in ways that asks questions, instead of answers them. Use relationships to other events, rather than exact times. Leave it out, if it doesn’t help the story. Think about what the book is, then do the research. Do you need to show the event happening or can you just tell the reader that the event happened and had an outcome? Sometimes, you don’t want to go there. Postpone that decision until you need it! Be aware of the uncanny valley of worldbuilding — far off, skip the details, it’s okay, we got the broad strokes. Too close, too many details, and suddenly readers start asking questions. Don’t fall into that valley! Watch out for the super-detailed realistic piece that makes everything else look fake. Focus on what you actually need to keep the story from falling apart. Avoid worldbuilding details that would ruin the story.
[Mary Robinette] Season 14, Episode 35.
[Brandon] This is Writing Excuses, What You Leave Out.
[Mary Robinette] 15 minutes long.
[Dan] Because you’re in a hurry.
[Howard] That probably wasn’t what I was supposed to leave out, but go ahead.
[Brandon] I’m Brandon.
[Mary Robinette] I’m Mary Robinette.
[Dan] We all just sat there, going, “What is he? Oh!”
[Mary Robinette] And you are?
[Dan] I’m Dan, I guess.
[Howard] And I’m Howard. And unapologetic.
[Brandon] All right. What you leave out.
[Mary Robinette] [garbled not amused]
[Brandon] So when I teach my students about this topic, one of the things I mention is when I was a newer writer, one of the things I got told frequently is that you want to, in worldbuilding, worldbuild a ton. But not put all of it in. Put enough of it in that the reader… You’re indicating to the reader that it’s like an iceberg, right? You can see the tip and you can see that there is so much more beneath. The more I became a published writer, the more I worked in it, the more I realized that that was… not a fantasy, but perhaps people in the business making it sound a little more grandiose than it is. Because most people I know do not worldbuild the entire iceberg and then show you the tip. What they do is they worldbuild the tip, and then they find a way to worldbuild a hollow iceberg that makes you think that there is the rest…
[Brandon] Underneath there. The goal in worldbuilding is not to do everything, just to do as little as you can and still look like you’ve done everything.
[Howard] Two nights ago, I was watching the special features for the movie Deepwater Horizon, for that film. They built an 85% scale oil rig over a little 3 foot deep pond. The reason they did it was so that when the actors were outside up high, shooting scenes, the actors are reacting as if they are outside and up high. They could have done the whole thing green screen, but they didn’t. They needed that level of verisimilitude. Then there was this point where the VFX guy said, “So, we didn’t actually build the whole oil rig. We only built the front.” You see this scene where the helicopter is coming in and the camera has panned around the oil rig and it is just… Like 25%, 20% of the oil rig. Then the VFX says, “This is what we had to build,” and throws all the other stuff in. After hearing how much time they spent building 20% of the oil rig for verisimilitude, the peace that they needed, this iceberg thing totally makes sense. Build the piece that’s required for verisimilitude. Drill all the way down on that. Then fix the rest in post.
[Brandon] So, how do we apply this to our worldbuilding? What do you guys do when you are worldbuilding? How do you give this indication that there’s more underneath there? How do you decide what to leave out of your story? How do you decide what not to worldbuild?
[Dan] So, following along with this set building metaphor here, I remember reading an early interview with Gene Roddenberry when they were doing the original Star Trek series. He said that he wanted to have an engine room, and they weren’t going to build him one, until he put that scene into the pilot episode. He’s like, “Look, well, we have to have a scene here. I’m sorry, there’s no way around it.” So they gave him an engineering. What I do when I’m building my worlds and planning my books is I figure out, “Well, where are my scenes set? Where do I want those scenes to be set?” Am I going to be talking enough about main engineering, for example, that I need to figure out what it looks like and where it is and how it works, or is my story going to focus on some other thing? So they didn’t build the entire, or even 20%, of the Starship Enterprise. They built a bridge and an engineering room and a transporter room, and that’s kind of it. Maybe some hallways. Because that’s where they knew their story was going to take place. So I try to figure out what am I focusing on, what am I going to be using the most, and that’s what I focus on.
[Mary Robinette] I’m very much the same way. I really only worry about the things that my characters are going to be directly interacting with. I want to make sure that I understand enough of how they interact, of how it works, so that the interaction makes sense. But, like, when we move through our daily life, we interact with a lot of stuff that… There’s a number of houses that you passed on the street and you have no idea what’s in those houses. But they’re still houses. You go to Disneyland. You don’t actually know what it takes to make Disneyland work. It’s just the front facing stuff. So one of the things that I do is that I think about the pieces that my character is going to have that direct interaction with, like you were talking about. One of the ones that I find works really well our past events. Referring to things… Usually these are things that I have no idea of what they actually are. But instead of saying, “Well, this happened in 1457.” Like, I don’t actually want to figure out how long ago a thing happened. I don’t know. So I’ll say, “Well, it happened during the… Right after the battle of the seven red armies.” Everyone’s like, “Oh, well, the battle of the seven red armies.”
[Mary Robinette] Clearly, she spent all of this time thinking about that. What that’s done is it saved me from actually working out a timeline. Because I’ve… Now I can place the battle of the seven red armies anywhere I need to be.
[Dan] One of the things that that suggests to me is that you have given them the information in a way that asks more questions rather than answers them. That gives a gre… I mean, we know when it took place, but we know it based on a relation to an event rather than an exact number of years. In the audience’s mind, it’s not answering the question so much as it’s saying, “Don’t worry, I’ve got this. Also, here’s something else to worry about.”
[Brandon] Have you ever spent a lot of time in your worldbuilding before writing or during writing a story and then decided to leave that out of the story?
[Mary Robinette] Absolutely.
[Brandon] When, why, and what made you make that decision?
[Mary Robinette] In the Glamorous Histories, for Without a Summer, I spent a great deal of time figuring out how Parliament worked in relationship to glamour, and what laws were being passed and not passed, and got into the novel and realized that that entire plot structure was completely irrelevant. I like knew… I had spent all of this research on this one particular historical figure who never appears in the novel now. It was basically, it just didn’t help the book. Chucked it. It was one of the things that made me realize that I really need to think about what the book is and then do the research. I will say that I approach my research now the same way that I… I mean, I approach my worldbuilding the same way that I approach my research, which is that all do like these broad strokes, but I only really drill down on the stuff that I actually need to.
[Brandon] I spent a lot of time in the Stormlight Archive before I was writing it, working on the writing systems. The glyphs that they were going to draw and things like this. I left that all out because once I actually wrote the book and I looked back at the stuff I’d done, I realized I’m not an artist.
[Brandon] Beyond that, I’m not an expert in languages and… I just hired that out. So I took all the stuff I did… I didn’t even give it to them. Because I’m like, “You know what, I’m going to use the text that I’ve written in the book.” I’m going to give this to the artist and I’m going to say, “What would you imagine this to be?” Isaac came up with stuff that was waaay better than any of the stuff that I had come up with. It kind of taught me, also, that maybe I should spend my effort where I know I’m going to be using it in the story, and then I can, after the fact, I can hire some of these things out.
[Dan] Brandon, you and I just did this yesterday, actually, on the project we’re collaborating on. The Apocalypse Guard. We’ve been wrestling with this book for months now, and yesterday made the decision that kind of the main thing we need to do to fix it is to axe one of the magic systems.
[Dan] It was something very cool that we considered foundational to the story, but now that we’re looking at the book in its current form, it’s kind of beside the point.
[Brandon] It’s also the thing that is causing the biggest problem with the story, because where the story is spiraling out of control are all these scenes where I spent lengthy amounts of time talking about the worldbuilding and the history. Scenes that Dan cut out a lot of when he did his revision.
[Brandon] But the effect of it’s still there. It’s leading to this big confusing ending where I have… Do what I do, tie all these worldbuilding elements together. But in ways that were cool for those worldbuilding elements and don’t really work for the story.
[Brandon] It’s a point where we have to cut out… One of the things that is my signature is a magic system. Granted, we have multiples. So it’s still going to be cool. But it’s going to be a way better book if we just streamline.
[Howard] My approach here is often to ask where the line is between show versus tell. There are times in the story where it’s absolutely required for the reader, because it’s fun, because there’s emotional content, whatever, to show an event happening. Then there are times when all the reader needs is to know that the event happened and there was an outcome. So entire scenes will vanish from the writing, because what I needed to do, with the story needed, was for somebody to say, “Battle was fought. So-and-so won.” “Oh, really, that sounds terrible.” And off we go with the core story.
[Brandon] Let’s stop for our book of the week.
[Mary Robinette] All right. So our book of the week is Stealing Worlds by Karl Schroeder. I got to read this, an arc of it. It is fantastic. This is near future. It’s an Internet of Things. A young woman discovers that her father has been murdered. She thinks. Everyone else thinks that it was a… Just an accident. Then people start coming after her. How do you disappear when everything is connected? So it’s really, really cool. It feels like he has thought of everything. But the stuff that we’re actually seeing is just the stuff that she interacts with directly. It’s great worldbuilding, great characterization. I mean, it’s a really good book. It also happens to illustrate some of these points.
[Brandon] Excellent. That was Stealing Worlds by Karl Schroeder.
[Brandon] So, we’ve talked about worldbuilding elements that we cut out. Are there ever things that you have decided even before you launch into the book, you’re like, “I’m just not going to touch that. I’m not going to go that direction with the worldbuilding.” Things that you just… Why have you done this?
[Mary Robinette] Oh, like in the Lady Astronaut books, I very carefully do not talk about what the rocket engine is that is driving this ship to Mars. I like really carefully do not talk about that. Because of the amount of research that I was going to have to do. But also, my character is not a rocket engineer. Right? She pilots things. She needs to know how to pilot things, and she does math. So, she needs to do those things. But I did not need to know how the rocket engine worked. And as soon as I worked on figuring that out, that was going to lock me into certain decisions. Like, if I decide that it is atomic oxygen, that locks me into one line of technology. If I decide that it is nuclear, that locks me into another line of technology. Because I don’t know what subsequent books are going to need, I decided to not make that decision and to leave room for it to be any of those things, and just… I establish some trust with the reader early on, so that I can just… Like, just get in there and…
[Brandon] You know…
[Mary Robinette] It’s like, they’re going to Mars. Obviously, they’ve solved how they get there.
[Brandon] I had a conversation with this… About this same topic with a writer that I know… That we were kind of brainstorming on some worldbuilding and things. The way I presented it as there’s like an uncanny valley of worldbuilding where at a certain point, it’s far off, and you’re leaving out the right details from what we’re doing so that nobody starts to question really how it works. Like, if you don’t do enough, people are confused and you start to lose them. You do the right amount, and people are willing to take your word on it. They suspend their disbelief, they accept the worldbuilding, it feels really logical to them, you’ve got the couple of corner cases that they would assume. Then there’s a stage where you start explaining it so much that the rational part of their brain kicks in and says, “Well, wait a minute. This and this and this and this,” and you start to hit this sort of uncanny valley where suddenly you lose them. They aren’t willing to suspend their disbelief anymore. That can be a really fine balance to walk.
[Mary Robinette] We have this problem in theater, with… All the time. Where you’ve got a set, and if you go very minimalist with it, you’re asking the audience to be engaged. You go too minimalist with some shows, and everything falls apart. But if you’ve got like a set where everything looks really nice, and then there’s this one piece that is hyper realistic, everything else in the story feels just awful. Beauty and the Beast, the animation… When they had… That was the first stuff of the computer animation…
[Dan] They introduced CG in the ballroom scene.
[Mary Robinette] The ballroom scene looks… It looks wrong, because it is more rendered than everything else. Then everything else starts to look false.
[Dan] I did a black box production of Assassins in college. It was all just super minimal sets, but we had a super realistic like rolltop desk, and it just… It looked terrible. Because it made the rest of the show looked terrible.
[Mary Robinette] One of my favorite pieces of set design that I ever did… This is a side tangent, but a good example. A friend of mine called me on a… On Monday and said, “We had a reading this weekend and are set designer did not show up with the set. I have just found out that she has skipped town with all of the money which she has spent on drugs. We open on Friday. Help me. I have $75.”
[Mary Robinette] So I’m like, “Okay.” We sat down and we talked about what are the things that have to be on stage or the show will fall apart. It was a tree, the moon, and a wall. That was basically it. So I bought some foamcore, and I got some paint, and I did this dry brush minimalist New Yorker style thing of a tree, a moon, and the wall. I think I gave him a chair, too. As a bonus.
[Dan] ‘Cause you’re a benevolent god.
[Howard] You had eight dollars left.
[Mary Robinette] I still had eight dollars. I had to get paid out of that $75, you know. So I… But we stripped it down to what you actually need or the show will fall apart. When the review came out, it raved about the minimalist design and delicate ethereal touches of the set. Meanwhile, in the program, I am listed as scene proctologist, because I pulled that set out of my ass.
[Mary Robinette] So, point being, just look at the worldbuilding details that you need to keep the show from falling apart.
[Dan] Well, it can also be helpful to look at the worldbuilding details that would ruin things. When I did my cyberpunk series, I specifically avoided artificial intelligence. There’s algorithms, there’s swarm intelligence, but there is no self-aware thing because that is a singularity that I was not prepared to deal with. So, that’s not in the story, it’s not a possible technology in that world.
[Brandon] This story of Mary Robinette’s actually leads us really well into our homework. Which Howard is going to give us.
[Howard] Yup. I want you to take your worldbuilding slider and I want you to pull it all the way to zero for one of your chapters. Take a chapter that’s got some worldbuilding exposition in it, that’s got some cues about what’s going on in your world that are deepening things, and pull all those out. Leave yourself with zero worldbuilding. Have a look at that chapter and see which elements of the story fail and which elements of the story still work. This is not so that you can tell yourself that you don’t need to worldbuild. This is so you can tell yourself… What the…
[Dan] I need a tree and a moon and a wall…
[Howard] I need a tree and a moon and a wall, and I will give myself a chair.
[Mary Robinette] As a bonus, in the liner notes, I’m going to give you a copy of the first scene of Shades of Milk and Honey in which I have done this exercise. So I have stripped out everything that I identified as exposition. I have to say, that scene is a mess.
[Brandon] This has been Writing Excuses. You’re out of excuses, now go write.