swan_tower | Entries tagged with how i do it

I think I've suddenly become an evangelist for figures of speech.

During a recent poetry challenge in the Codex Writers' Group, someone recommended two books on the topic: The Elements of Eloquence: Secrets of the Perfect Turn of Phrase by Mark Forsyth, and Figures of Speech: 60 Ways to Turn a Phrase by Arthur Quinn. I found both delightfully readable, in their different stylistic ways, and also they convinced me of what Forsyth argues early on, which is that it's a shame we've almost completely stopped teaching these things. We haven't stopped using them; we're just doing so more randomly, on instinct, without knowing what tools are in our hands.

What do I mean when I say "figures of speech"? The list is eighty-seven miles long, and even people who study this topic don't always agree on which term applies where. But I like Quinn's attempt at a general definition, which is simply "an intended deviation from ordinary usage." A few types are commonly recognized, like alliteration or metaphor; a few others I recall cropping up in my English classes, like synecdoche (using part of a thing to refer to a whole: "get your ass over here" presumably summons the whole body, not just the posterior). One or two I actually learned in Latin class instead -- that being a language that can go to town on chiasmus (mirrored structure) because it doesn't rely on word order to make sense of a sentence. ("Ask not what your country can do for you, but what you can do for your country": English can do it, too, just a bit more loosely.) Others were wholly new to me -- but only in the sense that I didn't know there was a name for that, not that I'd never heard it in action. Things like anadiplosis (repeating the end of one clause at the beginning of the next: "Fear leads to anger. Anger leads to hate. Hate leads to suffering.") or anastrophe (placing an adjective after the noun it modifies: "the hero victorious" or "treason, pure and simple")*.

*Before you comment to say I'm using any of these terms wrong, refer to the above comment about specialists disagreeing. That anastrophe might be hyperbaton instead, or maybe anastrophe refers to more than just that one type of rearranging, or or or. Whatever.

Quinn's book is the older one (written in the early '80s), and something like two-thirds of his examples are from Shakespeare or the Bible. On this front I have to applaud Forsyth more energetically, because he proves his point about how these things aren't irrelevant to modern English by quoting examples from sources like Katy Perry or Sting. (The chorus of "Hot n Cold" demonstrates antithesis; the verses of "Every Breath You Take" are periodic sentences, i.e. they build tension by stringing you out for a long time before delivering the necessary grammatical closure.) And when you get down to it, a ton of what the internet has done to the English language actually falls into some of these categories; the intentionally wrong grammar of "I can haz cheeseburger" is enallage at work -- not that most of us would call it that.

But Quinn delivers an excellent argument for why it's worth taking some time to study these things. He doesn't think there's much value in memorizing a long list of technical terms or arguing over whether a certain line qualifies as an example -- which, of course, is how this stuff often used to be taught, back when it was. Instead he says, "The figures have done their work when they have made richer the choices [the writer] perceives." And that's why I've kind of turned into an evangelist for this idea: as I read both books, I kept on recognizing what they were describing in my own writing, or in the memorable lines of others, and it heightened my awareness of how I can use these tools more deliberately. Both authors point out that sentiments which might seem commonplace if phrased directly acquire impact when phrased more artfully; "there's no there there" is catchier than "Nothing ever happens there," and "Bond. James Bond." took a name Fleming selected to be as dull as possible and made it iconic. And it brought home to me why there's a type of free verse I find completely uninteresting, because it uses none of these things: the author has a thought, says it, and is done, without any intended deviations from ordinary usage apart from some line breaks. At that point, the poem lives or dies entirely on the power of its idea, and most of the ones I bounce off aren't saying anything particularly profound.

So, yeah. I'm kinda burbling about a new obsession here, and no doubt several of you are giving me a sideways look of "ummm, okay then." But if you find this at all interesting, then I recommend both books as entertaining and accessible entry points to the wild jungle of two thousand years of people disagreeing over their terms.

(originally posted at Swan Tower: https://is.gd/08rQSn)

Last week, on a Tuesday when absolutely nothing else was happening at all, I put out a little book -- and I do mean little. It's right there in the title: The Writer's Little Book of Naming, an 11K-ish headfirst dive into the sociocultural side of naming -- particularly for people, but also a bit for places and things. It looks less at the conlang questions of phonetics and such, more at the ways names can reflect culture and, in so doing, help reinforce and deepen other aspects of worldbuilding.

This is actually the first installment of what I intend to be an irregular series, because it occurred to me one day that ebooks make it possible to assemble works on fairly specialized topics of craft -- the kinds of topics that can't really support an entire print volume, and which appeal to a niche market of writers, but dammit, I want to write about them, so here goes. I've got about six of these in mind thus far, so I'll update as they make it out into the world!

(FYI, it is currently available only through Book View Cafe, the publisher.)

I've spent the past several days (and have several more to come) gear-shifting between three radically different writing projects. On the one hand, I'm taking this approach because I know my brain can't just buckle down and slam all the way through one of them in a concentrated go; eventually it starts emitting steam and high-pitched whistles, and then I have to stop or switch to something else. On the other hand, that means I'm putting a different sort of strain on it, by asking it to get into a totally different mode on very short notice.

Thank god for the tactics I developed years ago.

It started out as a way to get myself into the headspace of a novel on days when I didn't want to write. Well, no, that's a lie; it started out as an accident: me being obsessed with a ten-minute trance remix of a particular song and listening to it on loop while I happened to be writing what eventually turned into Lies and Prophecy. But it became that thing I just said, and so I got in the habit of associating particular music with particular books. These days it's more often whole playlists rather than single songs; the former is slightly less insanity-inducing than the latter, but also (if we're being honest) a bit less effective.

This helps SO MUCH when I have to do this kind of gear-shifting. Even though two of the projects are new enough and small enough that they don't actually have associated music, I picked out an album in one case, a genre playlist in the other, and when I'm done with A and it's time for B, I change the music. And it helps. My brain goes, "Oh, techno? I absolutely cannot think about Previous Project with that going on. What else is on offer?" And then I open up the file for Next Project and we're off.

I'm not claiming it's foolproof. Also, not everyone can write to music (it's worth noting that the vast majority of what I listen to is either instrumental or in languages I don't speak well enough to be distracted by), so it's not a tactic that can work for everybody. Possibly you could sub in things other than music, like beverages or sitting in different parts of the house, though I think those would be weaker insofar as they're less likely to evoke particular genres, settings, and moods. But if you can do this: hoo boy does it help.

(originally posted at Swan Tower: https://is.gd/zm2fHq)

This is as much for my own records as anything else, but it might be useful to others, so I'm posting it here.

When preparing a list of names and terms to send your copy-editor, it helps a lot to do the following.

Open the file in a different word processor from the one you normally use. Doesn't matter which one; it just needs to be a program that will flag every word it doesn't recognize. (If you aren't the sort of person who normally teaches your usual program all the major names, you can skip this step.)
Start skimming, looking for those flags. When you find a name or term specific to your novel, check the spelling, put it in a list for the CE (helps to sort by characters, locations, world terms, etc.), then add that word to the program's dictionary so the flags go away. Note that you'll probably have to add it again for the possessive, though the CE list won't need that specified. (Plurals, however, might need to be mentioned on the CE list.)
If that word shows up again, it might be misspelled, or your program might do things like say "ACK THERE'S PUNCTUATION AFTER IT NOW IS THAT A NEW WORD???" If it's spelled correctly, add it to the dictionary again so the program will stop yelling at you. Over time, this will mean that the number of flags you have to scan for goes way down.
When you encounter things that aren't novel-specific terms, like compound words or variant spellings, add them to the CE list if you definitely want them a certain way (you'll change my strong preterites over my dead body); otherwise, add them to the program's dictionary so you won't have to look at them again, and leave finessing things like hyphens to the CE and their style guide.
By the way, you'll also catch a lot of typos.
Finish going through the manuscript. Hey presto, you have now -- probably -- found all the words you need to list for your CE.
Might also do your CE a favor by adding notes about who people are, what they look like, where in the city certain buildings are located, and other basic details of continuity. This doubles as a favor to yourself later on, if you're writing a series!

If you are writing a series:

For book two, open your CE list from book one. Add a 1 in front of every term, then search on the non-obvious ones to see if they're used again in book two. If so, add a 2 as well, so you get things like "1,2 Ren -- half-Vraszenian con artist" (etc).
Save this as your master CE list for the series.
Repeat the above process for adding to the program dictionary and CE list. Anything new goes in with a 2 in front of it, e.g. "2 Mede Galbiondi -- suitor with a strong arm."
Save a new copy as your CE list for book two. Then delete every item that has only a 1, i.e. doesn't appear in book two.
Delete the numbers from in front of your remaining items so your CE doesn't wonder what the heck those are for.
Rinse and repeat for book 3 et sequelae, starting with the master list and searching to see which existing names and terms show up again.

I learned at the CE stage of The Liar's Knot to create the master list. I learned at the CE stage of Labyrinth's Heart that I should have been numbering all the items so I'd know whether a lack of numbers meant that it showed up in both of the first two books, or that it got introduced for the first time in book two. (That part probably isn't necessary, but my brain wants it.)

You could probably just inflict the whole master list on your copy-editor each time around, but given how much these names and terms pile up, I feel like it's better to give them only the relevant selection, without cruft leftover from previous books.

Or, I mean -- you don't even have to make a list. I didn't for my first I don't know how many novels, until I heard this was a helpful thing to do. It's especially useful to you as a writer when you're writing more than one book, because of how it can double as a minimalist series bible. And if you're going to make such a list, this workflow minimizes the amount of mental effort that goes into finding all the names and terms. It's still time-consuming, but it isn't hard.

(originally posted at Swan Tower: https://is.gd/FQ5Zy3)

Having made a general typology of series (with a lot of good comments on the DW version of that post in particular, unpacking the various gradations between what I called the Setting Series and the Cast Series -- I'll include those when I make a more polished version of these posts), I want to start chewing on "so how does one write one of these things, anyway?"

I mean, you can just start typing and not stop until you have lots of books. But I don't recommend just charging in blindfolded like that. :-P

Obligatory Disclaimer: prescriptive writing advice is a mug's game, since somebody will always come along with an example of people not doing it that way and ending up fine. This is me talking about what I suspect may be helpful, not what is required. It's what I would say to somebody who feels adrift and needs some direction. And in that vein, I welcome comments about how other people view this process.

The first thing I would say to do is, consider what type of series you're writing. Not in the sense of "oh, I must fit neatly into one of those pithily-named boxes or I'm Doing It Wrong," but in the sense of knowing how much interconnection you want. This will partly be determined by individual preference (some people love strong arcs; some people don't), and partly by the nature of what you're writing: the conventions of the romance genre are not well set up for doing a long metaplot where the relationship of your two leads gets developed over the course of three or five books, while mystery, as mentioned before, is ideally suited to a more episodic structure, with variable amounts of growth and arc-ness in the background.

What type of series you're writing can shift over time (and one of the things I want to chew on in a future post is how to make changes mid-stream), but anecdotally, I feel like the direction of the shift will almost always be toward a greater degree of interconnection. An Episodic Growth series may sprout an arc plot after a while, especially if the writer is looking to wrap it up with some kind of grand finale. A series that told a relatively self-contained story in its first volume may shift to being a Metaplot Series once the writer has the certainty of being contracted for more books; you see this a lot with trilogies, e.g. the original Star Wars going from "yay, battle won; medals for everybody!" at the end of its first installment to "shocking revelations, severed hands, and unfinished business" at the end of its second. Shifting in the other direction is odd: a series with an arc is unlikely to ditch that and go purely episodic, unless it's headed toward a new arc (e.g. from season to season in a TV show). When a jump happens all the way to a Cast or Setting Series, what's usually going on is that one series has been wrapped up and the author wants to start another -- this being why the Setting type is so often a super-set of the other types.

Once you know what you're writing . . . then what?

My main thought on this front leans on that typology: The closer your series is toward the discrete end of the spectrum, the more you benefit from having a foundation. The closer it is toward the interlinked end of the spectrum, the more you benefit from having a target.

Foundation first. One thing I admire about the Dragon Age franchise is that its creators did a very intelligent job of creating a setting that could support the franchise -- a world with lots of narrative toys in it that could be used to generate plots. In the first game, the plot is that horrible monsters from underground are attacking the world at the direction of a draconic archdemon and spreading their corruption to the surface; your job is to stop that invasion and defeat the archdemon. But along the way to doing that, you encounter all kinds of other conflicts in the world. Mages are susceptible to demons, and as a result are supposed to be monitored and controlled by templars, with major tensions between those two groups. Elves once had an amazing civilization that got smashed and now they're an oppressed underclass in the human world. The main religion has heresies. The qunari are invading the continent, many many miles away from where the story takes place. Etc. These things help generate side quests for the game, but they're waaaaay larger than they need to be for that purpose.

Because those things exist, though, it's easy to tell other Dragon Age stories that have nothing to do with darkspawn and archdemons and Blights. The second game focuses on the mage/templar conflict, with a substantial contribution by the qunari invasion. The third game digs into the elven stuff, as does one of the novels. It's extremely fertile soil for many kinds of stories -- which is exactly what you want, if you're trying to create a franchise.

Or let's say you're writing a Cast Series in the romance model. Here your foundation is based in character: if you've got five brothers you intend to marry off, then you want to spend at least a little time thinking about their personalities and their conflicts, so that you put yourself in a good position to tell a new story about each of them. You don't have to outline every single one of those books, or even necessarily know who they're going to wind up marrying; their eventual partners may or may not be on stage in book one. But the brothers probably are, and knowing a bit about them now means that 1) they can be vivid personalities from the start and 2) you won't paint yourself into a corner later on because you've run out of interesting ideas for them.

Even in the Episodic Growth type of series, there's foundational work to be done. If you're writing a procedural series about a detective, with the intent that the detective's personal life will change over time, then it's beneficial to set her up with some conflicts that can generate good side plots throughout. Maybe she's got a drinking problem because of the trauma she's already been through, which she's trying to keep hidden. That means you can later make plot out of the problem getting worse, and her partner finding out, at which point she goes to rehab, but then later she falls off the wagon and her partner leaves her -- you've got books' worth of material right there. But you aren't necessarily committing to that path; maybe she goes to rehab, but then her partner cheats on her, so she ditches the partner and spirals back down into alcoholism. You can also say she's estranged from her family, which sets you up to have a relative show up eventually, with plot ensuing. Seeding two or three such things at the outset gives you something to work with in the longer term, separate from the individual plots of the episodes.

The reason you want to think about this ahead of time is that it reduces the need to pull something out of your ear later on -- and the risk of the reader noticing. If Dragon Age hadn't set up the mage/templar conflict at the outset but suddenly in the second game OMG mages are totally dangerous and can become demonic abominations . . . the player would be thoroughly justified in asking why this didn't come up before, when they were running around with mages in their party. If one of the romantic brothers is a colorless non-entity who unexpectedly turns into a notorious rake because it's time for his book, your reader will side-eye you; if the series is going well and so, uh, there are actually six brothers instead of five, that's going to look like exactly the cheap trick it is. (You can try to say the extra one never got mentioned because he's the bastard half-brother nobody talks about . . . but you've still missed your opportunity to set that up and make the reader crave his story.) And we've probably all had the irritating experience of following an episodic series where the writers abruptly graft a heretofore unsuspected problem onto a character because they need to generate new plot.

But if you're coming at things from the interlinked end of the spectrum, I think it's helpful to approach it with a different mindset. I've already been wordy enough, though, that I'll save that for a later post.

(originally posted at Swan Tower: https://is.gd/FtjzOu)

I've had discussions with other writers about how there's tons of advice out there on writing novels, but very little on writing series.

File this one under "stuff I know how to do, but don't know how to articulate or explain." But this one will be less polished than the pieces I wrote on the structure of paragraphs, scenes, and chapters, because I'm really thinking out loud as I go here.

Step one, I think, is to take a look at what a series is. A set of interconnected books, okay. But there are ways and ways of connecting things, and they're not all going to operate the same. After chewing on this for a while, I've decided that you can very roughly sort different types of series into a spectrum from discrete to linked (with two semi-outliers that I'll note as we pass them.) So:

The Non-Series

At the absolute discrete end, you've got books whose only connection is that a single author wrote them. Not actually a series; 'nuff said.

The Setting Series

In this type, the connection between the books is that they take place in a single setting, but otherwise they share no connection of character or plot. (They may not even share authors.) I'm having trouble thinking of any pure examples of this; most often this tends to be a superset of other series, e.g. Discworld or Valdemar being settings that contain both stand-alone novels and series within them, or a shared world like the Forgotten Realms. If you can think of an example that is purely stand-alone novels, whether written by the same author or different ones, let me know. (I think it would need at least three books to serve as a good example; two books in the same setting is a series by the most technical definition, but I'd like something stronger.)

The Cast Series

This is the type of series you commonly find in romance, where each book follows a different set of protagonists and a different plot, but characters from one book appear in another. (Romance often sets this up by presenting you with a group in the first book, e.g. a set of siblings, with the implicit promise that you'll get to see each of them get their own story eventually.) These naturally share a setting as well.

*The Reset Button Series

As the asterisk indicates, I think this one's an outlier. It's the Nancy Drew model: each book shares a setting and a core cast with all the others, but in between books the slate gets wiped clean, which means they have less plot continuity than the Cast Series. Nancy will always be eighteen; Ned will never graduate from college. I'm not sure this is very popular anymore, except maybe in children's fiction -- and maybe not even there?

The Episodic Growth Series

Closely akin to the Nancy Drew model, this has a core cast and a new plot with each installment, but there's no reset button. As a result, change and growth do happen over time. You see this a lot in mystery novels and police procedural TV shows, because it's very well-suited to those genres: each installment starts with a crime and ends with the crime being solved, while in the background there might might be some ongoing character-based subplot about the detective's marriage falling apart or whatever.

The Episodic Arc Series

This one is a hybrid between the previous and the subsequent types. It has self-contained episodic plots, especially early on, but there's also a longer-term metaplot that those episodes may be helping to set up, and the episodic structure tends to fall away toward the end. Examples include Harry Potter and each season of Buffy the Vampire Slayer (and yes, I realize the creators of both those works are not exactly looking great right now, but they're well-known illustrations of the model). Many trilogies feel at least a bit like this, because it's sensible from a business standpoint to write a more or less stand-alone novel that can serve as the foundation for the later two installments.

*The Perpetual Motion Series

Our other outlier, which I think I've only seen in soap operas on TV. Here there can be many arcs going at once, such that while an individual plot may end, the series as a whole doesn't (until it gets canceled). This would be an extraordinarily hard trick to pull off in traditional novel publishing, I suspect, though it could work in indie.

The Metaplot Series

Here there's no real attempt to wrap up a self-contained plot in any particular installment. From the start, you know you're getting a long-term story, and unlike that trilogy approach I described above, the first volume doesn't feel like it could stand on its own. A Song of Ice and Fire is a prominent example of this, along with TV shows like Lost.

The Single Book

And to cap off the other end, we have our other form of non-series: a single novel that just happens to have been published in multiple volumes, i.e. The Lord of the Rings. The difference between this and the Metaplot Series is that in theory the author of the latter type gives each book its own satisfying structure, even if that structure doesn't end in resolution; the author of the Single Book non-series just whacks it apart at the necessary intervals.

I think that covers the whole gamut. Obviously some things are going to straddle the divisions, because no system of categorization is ever perfect; the goal here is to distinguish what shifts of interconnection happen along the way, rather than to make clean boxes that absolutely everything will fit neatly into. And series can change over the course of their lifetime, e.g. what the author intended to be Episodic Growth sprouts an arc plot along the way. I'll chew more on those bits of the concept later. But for right now, I think this is a decent framework? Is there anything significant I'm missing?

(originally posted at Swan Tower: https://is.gd/KmOFjZ)

I'm noodling around with an idea that I think will be a novella, and part of that noodling involves thinking about novellas in general.

I don't have the world's best grasp on how to pace a story of this size. I've written five of them, but two weren't planned that way -- both Deeds of Men and Dancing the Warrior were me saying "well, let's write this idea and see how many words I end up with" -- and really, five isn't all that many in general. Nor do I think I'm alone in being uncertain about how best to structure such things: a lot of the novellas I've read feel like they aren't paced quite right, going too slow in some places, too fast in others. I speculated to a writer-friend on a forum that it's because novellas were kind of a dead zone in SF/F for a long time (few good ways to publish them, so very few people writing them), and we can't look to the novellas of the more distant past for much guidance, because our expectations of storytelling have changed. We're sort of reinventing the wheel, now with suspension and treads and spinning rims.

Whether I'm right about that or not, the fact remains that novellas feel like terra barely cognita to me. Plus I'm not the kind of writer with much in the way of overt understanding of pacing anyway; what I do, I tend to do on instinct. I know plenty of writers who love making use of beat sheets and the like, which map out what kinds of events should happen when in a novel, but those are deadly to my process. So even if you had a beat sheet for a novella, I wouldn't get much use out of it.

But the other day I realized that I do have one useful framework for thinking about this. I need to ask myself: is what I'm writing more like a short story, or more like a novel?

With a novel, I usually have a couple of set points I vaguely map out ahead of time, pegging them to what feels like the right moment in the story -- most often either the 1/3 and 2/3 marks, or 1/4, 1/2, and 3/4. The Night Parade of 100 Demons is a thirds novel; The Mask of Mirrors is a quarters novel. Sometimes I can tell you why; In Ashes Lie is done in quarters because the Great Fire of London burned for four days, and I knew I wanted to interleave two timelines (the fire and what led up to it), so that meant breaking the preceding history into four chunks. Sometimes I have no idea. Any novel I break into an odd number of segments often winds up with a midpoint marker as well; that's true of both Night Parade and Midnight Never Come (which is in five parts because I wanted to echo the five-act structure of a Shakespearean play). I don't push too hard for them to wind up exactly on their marks, but since I have a general sense of how long I want the book to be, I use the set points to gauge how much complication and side plot should develop on the way to the next milestone.

That is not at all how I approach a short story. With those, it's never a matter of me wanting to place narrative turning-points at certain percentages of the way through the total. My short stories are usually built as a sequence of scenes: whether they get scene breaks between them or not, I know we need X to set things up, Y to develop them, and Z to conclude them. For highly variable quantities of X, Y, and Z, of course, and sometimes the structure is non-linear or whatever -- but it doesn't change the essential point that I know what needs to happen, and the quantity of words needed to do that properly determines how long the story is. Which nearly the opposite of the novels, where I've got a ballpark target for length and a few key fragments of what's going to happen, with the bricks being filled in as I feel my way through the story.

(Obligatory disclaimer: writing it out this way makes it all sound much tidier than it is in reality. For example, tons of my short stories start out with me having no idea where I'm going with my shiny new idea. Then they sit around until I've figured out enough of the remainder to write the rest. Sometimes this takes years.)

So where do novellas fit into this? As soon as I asked myself "should I approach them like a short story or like a novel," the answer was obvious. They're ickle novels, not gigantor short fiction. I'm not going to be able to see the full sequence of scenes ahead of time, no matter how long I let it sit. Which means the thing to do is to find myself a couple of fixed points and then decide where they should go. This feels like a thirds story to me: at roughly the one-third mark, the protagonist will succeed in getting E and G out of the situation they're in, and then at the two-thirds mark they'll . . . either get to where she promised to take them (only to find more complications there), if I decide that's the way the story is headed, or they'll abandon that goal and do something else. I don't know which, but I don't have to. It's enough for me to say, okay, something like 8-12K of "getting them out of their situation" plot, then another 8-12K of "difficulties and developments along the complete lack of road" plot. Writing the latter will tell me what's going to happen at the two-thirds mark -- or if it doesn't, then I'll let it sit for a while. (That happens sometimes with novels, too. A Natural History of Dragons stalled out one-third done for several years.)

I can't swear this is going to produce good results, because I haven't tried it yet. But it feels right, y'know? It feels like an approach that will help me thread the Goldilocks needle of too much or too little plot for the space available. I know when the narrative is going to change its trajectory, so now it's just a matter of feeling my way through the smaller conflicts and alterations before then.

I will report back!

(originally posted at Swan Tower: https://is.gd/bXMS44)

More and more, I feel like what I wrestle with in plotting novels is not figuring out what should happen, but what order it should happen in.

This is almost certainly prominent in my mind right now because of the Rook and Rose books, which, with their multiple points of view and interweaving strands of plot, pose more complex sequencing challenges than a single-pov novel with a more linear narrative. The same is true to a lesser extent, of The Night Parade of 100 Demons, where the two protagonists are pursuing the same main objective, but with side plots lacing through and around it. Even in a straightforward book, though, sequencing can be a question, because I also wrestle with this on the level of an individual scene. When more than one thing needs to happen, which deserves the highlight provided by being positioned at a key spot? Which one leads more naturally into the other? Are things easier or harder for the characters if they go in a certain order? Does that create an echo or a contrast with the way things went elsewhere? Has there been too much of the (metaphorical) color yellow for a while, and we need some blue to break it up before we go back to that?

I'll use the first Star Wars movie as an example. In the early part of that film, the sequence goes: Luke acquires droids; Luke follows R2-D2 and meets Obi-Wan Kenobi; Obi-Wan invites Luke to go with him and learn to use the Force; Luke declines; Luke goes home and discovers his aunt and uncle have been killed; Luke decides to go with Obi-Wan after all. You could, if you wished, change the sequencing such that the Stormtrooper attack comes earlier and Luke escapes with Threepio and Artoo. Then he goes to Obi-Wan for safety, learns all the stuff about Leia, and accepts Obi-Wan's offer the first time it's made, because by then he's already lost everything at home.

And that would work! It would just work differently. The structure would be less mythic (because Lucas was following Campbell's model in The Hero With a Thousand Faces, where the Call to Adventure is often refused at first), and it would radically change the tone of Luke's initial encounter with Obi-Wan, since then he'd be a fugitive grieving the loss of his family. It would also change your perception of Luke as a character: as the story is structured in reality, he dreams of getting off his podunk farm -- but when the chance to do that is offered to him on a platter, he turns it down because he's a good kid who doesn't want to leave his aunt and uncle in the lurch. If you remove the aunt and uncle before the choice appears, you never see that side of him; indeed, it won't feel like Luke is making much of a choice, because there's already no reason for him to refuse (and plenty of reason for him to agree). But that doesn't mean the other sequence doesn't work; it all depends on what effect you're trying to achieve. Not everything is aiming to be mythically structured, nor centered on a good kid with both dreams and a sense of responsibility.

So once you know what's going to happen in a story, there might still be decisions to make. Some things have to go in a certain order; Luke has no reason to visit Obi-Wan before he acquires the droids, and if they ran into each other for some reason, it would be an empty scene, with much less for them to talk about. (Movies in particular can't really afford empty scenes, but even novels shouldn't have them: if the sole reason you've got for an encounter is "to establish that this character exists," ask yourself if you can just wait until there's something to do with him.) But I think that for all but the most driven thriller plots, there's often wiggle room. If the blue bit of plot will provide your character with a safety net for the dangerous thing in the yellow bit, is it better to do the blue part first? Maybe yes, maybe no; it's more exciting if they don't have the safety net, but if the character is someone who simply would not risk doing the yellow thing without a fallback plan in place, then delaying the blue might seem like bad characterization for the sake of drama. Whether you're looking at the larger narrative arc or the flow of a single scene, it's all going to depend on the material at hand.

Which is why my novel-writing process increasingly features index cards with bits of plot scribbled on them. Those are very convenient for shuffling around on the floor, test-driving different sequences to get a feel for what will play best.

(originally posted at Swan Tower: https://is.gd/GKhjMi)

(This is the fourth post in a series about the craft consideration that goes into deciding where to put breaks between units of your story. Part I, Part II, Part III.)

As I said at the beginning, this whole series of posts sprang out of a conversation I was having with other writers about chapter length, which included some discussion of deciding where to start and end a chapter, i.e. where the breaks should come between them. After three posts mostly about other things, we at last come full circle back to the original question.

( Read more... )

(This is the third post in a series about the craft consideration that goes into deciding where to put breaks between units of your story. Part I, Part II, Part IV.)

The second post of this series looked at the ideas of attention and focus, and how those apply to the structure of a paragraph. Now let’s turn those same lenses onto scenes.

First, the notion that a unit asks you to sustain your attention until its over. Scenes don’t require the same degree of concentration from the reader as a paragraph; if you put a book down in the middle of a scene to go refill your water glass, you probably won’t have to start over at the beginning because you don’t remember where you left off. But ideally, a scene should hold the reader’s attention without pause, and not let them up for air until it’s done.

One of the ways it can do this is through unity. We no longer hold to Aristotle’s <https://en.wikipedia.org/wiki/classical_unities>classical unities as such, but in some ways the concept is still alive today at the scene level: we do generally expect unity of viewpoint, as I mentioned before, and we have a tendency to default to unity of location and time as well. When the characters shift location or a lot of time passes, we often insert a scene break to signal the transition and skip over the intervening gap.

But that isn’t the only way to handle those shifts. You can also use the narration itself to signal movement or the passage of time. How do you know which approach is better in a given situation?

( Read more... )

(This is the second post in a series about the craft consideration that goes into deciding where to put breaks between units of your story. Part I here; Part III here; Part IV here.)

In the first post of this series, I talked about the mechanics and pacing of where to break between paragraphs, scenes, and chapters. But “you have to start a new one under these conditions” and “merits and demerits of short vs. long” doesn’t get you very far; there are still enormous aesthetic decisions involved in where you choose to place your breaks.

(This is where I start flailing vaguely in the direction of articulating things I know, but have never tried to explain.)

As I said in that first post, I think this is largely a matter of regulating your reader’s attention. Unpacking that more, I think there are (at least) three aspects to this:

A unit, be it a paragraph, a scene, or a chapter, asks the reader to sustain their attention until it’s over. The intensity of that attention varies -- more for a paragraph; less as you go up the scale -- but if they’re going to look away, they should ideally do that when the unit ends, not partway through.
A unit is a way of signaling to the reader that there is a relationship between its component parts. Units whose component parts are unrelated are usually less effective -- and again, that’s most true at the paragraph level, and less so as you go up the scale.
Finally, a unit guides the reader’s attention to particular points of focus. This is primarily true at the beginning or end of the unit.

Because the operation of each of these things differs significantly between sizes of unit, let’s take them one at a time, starting with paragraphs.

( Read more... )

A discussion among my fellow writers of chapter length and where to break (or not) got me reflecting on how little writing advice there is for thinking about this -- and then from there I fell down a rabbit hole of realizing how even less advice there is for the sub-units below the chapter, the scene and the paragraph. (Or the higher-level units, the part or the book in a series . . . but that’s going to have to be a separate bit of pondering.)

This is stuff we’re apparently expected to learn by trial and error. You write stuff, and you notice -- somehow -- that breaking in certain places works better than others, and so you improve. Nobody ever really taught me how to think about these issues, beyond a few very basic mechanical points, and so as a consequence I’m not even sure how to articulate what it is that I do, even though I’m relatively pleased with how I’m doing it. This is the first in a series of posts that constitute an attempt to figure that out by talking through it out loud (so to speak), and I hope it will be of use to other people.

Note: what I have to say here is geared toward fiction writing, but certain aspects of it would apply to nonfiction as well, whether that be a blog post or an academic article.

Organizing it is a little bit hard, though, because I want to talk about all three of paragraphs, scenes, and chapters, and some of the points apply to all of them, but some don’t. Which means it’s not ideal to separate them, but it also isn’t ideal to tackle them all at once. I’m going to do a little from Column A, a little from Column B; I’ll start out with talking about the aspects where they’re the closely related, then break it up for where they diverge. Which also means this is going to be a multi-part discussion -- four parts in total, with one being posted each day. (Edited to add: Part II, Part III, Part IV.)

So with that context out of the way . . . in thinking about this, I’ve come around to the opinion that there are three major factors at play in how we decide to break up the units of our tale. Those are: mechanics, pacing, and attention. And of those three, I think attention is both the most subtle and the most important.

( Read more... )

A sentence I revised tonight got me reflecting on one of the tiny, subtle things about writing that’s really difficult to teach — mostly because it requires spending a disproportionate amount of time talking about something microscopic, and unpacking it with twenty or a hundred times the number of words involved. I almost never delve into this when I teach creative writing, just because it burns out my energy so fast.

Consider this line:

“I wouldn’t have invited you if all I wanted was a distraction.”

This is a completely ordinary sentence. Not super-memorable, but it’s doing what it needs to, and that’s fine. What I revised it to was this:

“If all I wanted was a distraction, I wouldn’t have invited you.”

Essentially the same sentence; I just swapped the order of the clauses.

Why? Because the important thing in that sentence is the implication that the invitation was issued for more than one reason. Putting the hypothetical after that dilutes the effect. So I rearranged the sentence to make the punch arrive at the end of the sentence.

Now, in reality a person might well choose the first phrasing. We often talk that way. But the job of dialogue is to create an effect, and while sometimes the desired effect is “the casual structures that mimic real speech,” in this case, that wasn’t the goal. There isn’t a clear-cut rule, though, that says “always put the most important thing at the end of the sentence” — sometimes you want it at the front instead. The actual rule is “pay attention to the rhythm of what you write, not just in aural terms, but in terms of where you’re placing the key elements, and make sure the arrangement directs the reader’s attention toward them, without less-critical elements getting in the way.”

Which is a lot more complicated and subjective. In fact, some of you may question the superiority of my second example over my first. Because it’s not just about the one sentence; it’s about the flow of the overall text. (Unfortunately, I can’t quote the surrounding material to you because SPOILERS AHOY.) And even when the whole is available, there can be disagreements over what works best. But when I read a story that’s competent but never quite comes to life, the problem is often (at least in part) at this level: the material is all there, but the sequencing undercuts its effect. Teaching that to someone, however, requires breaking out the red pen and rewriting sentence after sentence, with explanations for why. It’s a huge investment of time and effort, and in the end, the writer needs to develop their own instinct for how these rhythms work.

There are a lot of TV shows I try and just sort of drift away from, because they aren’t doing enough to hold my attention. The latest in this series is Black Lightning, which surprised me, because there are a number of things I like about its characters and its story. But in the end, its dialogue doesn’t have much of a particular element for which I can find no better term than “zing.”

Thanks, brain. “Zing.” That’s a real helpful way of describing it. >_<

Zing is not the same thing as witty banter — though many shows have mistaken the one for the other, and fill their scripts with dialogue that’s absolutely leaden in its attempt to be light. You can have zing in a deadly serious conversation (as Game of Thrones has proved). It’s a cousin, I think, of Mark Twain’s comment about the difference between the right word and the almost-right word being the difference between lightning and a lightning bug: it’s the lightning lines, the ones that leap off the page or the screen, the ones that don’t just get you from Narrative Point A to Narrative Point B but make the journey between them memorable. You see it in The Lion in Winter, which along with Twelve Angry Men made me wonder if this is a quality especially possessed by older stage plays — I haven’t seen enough older stage plays to be sure. At its apex, it’s the feeling that no line has been wasted or allowed to do the bare minimum of work. Think of The Princess Bride, and how many lines from that movie are quotable. It isn’t just because the lines themselves are good; it’s because there’s almost no flab in the script, every word simultaneously developing character and furthering the plot while also being entertaining.

Zing gets my attention, in a TV show or a movie or a book. Without it, my attention wanders a bit; I scrape a general sense of the story out of the mass of words used to tell it, but don’t engage on a moment-to-moment level. With it, I lose track of the world around me because I don’t want to miss anything in the tale. Zing makes me decide, before I’m two scenes into the first episode of a show, that I’ll give the second one a shot. Zing is what makes me plow through thousands of pages of Neal Stephenson making an utter hash of his plot, because he can describe a room in above a tavern on the seventeenth-century London Bridge in such riveting terms that I wind up reading it out loud twice, once to my husband and once to my sister.

I think this is what some people, when teaching the craft of writing, describe as “voice.” I’ve been known to rant about how I find that term completely unhelpful . . . but, well, here I am talking about “zing,” because my alternative is to wave my hands around in the air and make inarticulate noises. That thing. Over there. Do you see?

These days I’m reaching for it more in my own work, especially in one of the things I’m noodling around with right now. A character is hiding in a palace full of baroque decorations and complaining about the discomfort. There’s something jabbing into my back. No. There’s a carving jabbing into my back. No. There’s a gilded carving grinding into my kidney. Better. There’s a gilded figure of the South Wind imprinting itself on my left kidney. Better still.

Doing that for every sentence is exhausting. I have no idea how Stephenson keeps it up, especially while writing books that could double as foundation stones. But I suspect that, like many things in writing, after you’ve pushed at it for a while some parts of it just settle in as habit. I hope so, anyway, because I’m going to keep trying.

Mirrored from Swan Tower.

I can't say a lot about the work I do for Legend of the Five Rings because I signed an NDA. But the most recent round of brainstorming for a fiction has me reflecting on what this job is teaching me about making sure that the material I write pulls as much weight per word as possible, and I want to discuss that a little. So let's see what balance I can strike between specificity and deliberately vague generalities!

The context here is that I have a fairly strict word count for each of my fictions: 3000 words max if they're going into a pack, and 3000 with some wiggle room if they're being published on the website. That is . . . not a whole lot. And the story of L5R is so sprawling that even with a bunch of writers producing a bunch of fictions, making sure that everything gets mentioned and explored and moved forward means we can't afford to waste words. It isn't enough for a given fiction to do one thing; it needs to do at least two, more like three or four, as many as we can stuff in there at once. Ten pounds of story in a five-pound sack.

Take the one I've got on my plate right now. The original query from the person I work with Fantasy Flight Games was, "Are you willing to write a story about Character and Group? Something to flesh them out."

Me: "Sure! What do you think of Scenario?"

FFG: "Sounds good. Maybe you could work in how Character feels about Key Theme, and also expand a bit on Group's Main Focus."

Me: "I lean toward having Character feel this way about Key Theme, because that lets me make a contrast with Previously Mentioned Backstory Character. And for Group, maybe Side Character says XYZ -- that adds depth to their personality because of Probable Reader Interpretation. Heck, I could even put in Callback to Other Plot A, in a way that layers in some ambiguity."

FFG: "Great!"

Me: "OOOH. And -- just spitballing here -- but given the timing, what if we say that Side Character also has Information about Other Plot B, which of course they interpret in Particular Way?"

FFG: "Go for it. But maybe spin it a bit more to the left to emphasize Aspect."

Me: "Awesome. I'll have an outline for you shortly."

It could have just been a story about Character and Group. It probably would have been a perfectly fine story. But the more we can build up these elements, expanding on some things and contrasting with others, making callbacks to previous material and introducing points of linkage in all directions, the richer the fiction becomes.

Not all of this will stand out, of course. Sometimes the work the fiction is doing is fairly subterranean, and only somebody who's digging into the craft of it will notice that, for example, we're spinning that last bit to heighten a particular flavor. The overall effect is there, though, and in the long run it pays off: you can poll the readership and they'll agree that Character Q would never do a particular thing, without you ever saying that outright, because you've put enough data points on the table that they can extrapolate as needed. Things become three-dimensional; they feel interconnected. The world feels real.

In my novels I have a lot more room to work with, but it's still a good lesson to bear in mind. Why just have two characters converse with each other, when their conversation could also be making metaphorical allusions to something from earlier and enriching the reader's understanding of someone else not present for that scene? Why solve conflicts one at a time, when the solution could be taking out two problems, creating a third, and sending a fourth in an unexpected new direction? This is pretty standard advice for writing, but I feel like the level to which I'm doing it here is higher than usual, and rewardingly so.

Sustaining that over the long run is tough, of course. On the other hand, this is like a muscle: the more you use it, the stronger it will get. So I'll keep pumping narrative iron.

When writers talk about questions they get asked too often, "Where do you get your ideas?" is often high on the list.

Which is odd to me, because I've rarely been asked that.

"Where did you get the idea for this book?," sure. Got that one a lot with A Natural History of Dragons and the Memoirs of Lady Trent in general. But as a broad inquiry into my work as a writer, no. Still, it seems that other people do get asked about it frequently, so lately I've been pondering it, that I might be prepared when the question comes my way.

( Read more... )

yhlee recently posted about How to Write a Sizzling Synopsis by Bryan Cohen, which is a topic that's been on my mind lately. I can't swear that I'm a genius at cover copy -- what Cohen calls a synopsis; it's the stuff written on the back or inside flap of the book, or in the "description" field online -- but I actually enjoy writing it. And lately I've found myself even thinking of various works in progress from that angle, because figuring out what I would put into the cover copy helps me focus on what's core to the story, what I want to use to hook the reader.

Basic principles: you want the reader to know who your protagonist is and what conflict they face, and you want to do so in a fashion that's consistent with the overall mood, whether that's lighthearted or lyrical or grim. After that, you walk a tightrope between being specific enough to convey flavor and being general enough that you don't drown the reader in new information. The latter is especially tough in speculative fiction, where sometimes presenting the conflict is nigh-impossible without first explaining the world. (Ask me some time about trying to summarize the Varekai novellas. Or better yet, don't.) Writing cover copy requires you to develop your eye for what details are load-bearing (the text will make no sense without it), what details are beneficial (not necessary, but they add a lot), and what details are extraneous.

For novels, I often adhere to a three-paragraph approach. The first paragraph introduces the situation; the second introduces the problem; the third leaves the reader with a sense of momentum and/or tension, a clear awareness that you have shown them the tip of the iceberg, but there is much more to come. Yes, it's formulaic -- but formulas come into existence because they're good, reliable workhorses.

Since discussing this kind of thing goes better with examples, I'm going to dissect my own copy for Lies and Prophecy, because I can say exactly why I made the choices I did. (It's also my earliest effort, so not the best, but in some ways that makes it even more instructive.)

( Read more... )

I haven't said much here about my work on the current novel -- the one that's a followup to the Memoirs -- in part because it is so unlike the process of writing any other novel so far, I'm too busy figuring out what I'm doing to spare much attention for reporting in.

But hey, it's useful to talk about what happens when you write a Totally Different Kind of Book. So here goes.

( Read more... )

So if you need me, I'll be over here with a pile of mosaic pieces and a half-finished picture on the floor, trying to decide exactly where each tile should go.

I’m starting to think there are two kinds of research — or rather, a spectrum with two ends. Quite possibly it’s a more multi-directional spectrum than that, but there are two ends that seem particularly applicable to my life.

The first kind is reading for facts. This is the type of research I did all the time for the Onyx Court books: I’m writing about a specific thing, and so I need to know stuff about it. What route did Elizabeth I’s coronation procession take? Where were the imprisoned members of Parliament held after Pride’s Purge? When did somebody calculate the moment of perihelion for Halley’s Comet in 1759? What actions were taken by Fenian terrorists in the later Victorian period? This extends to more general questions; a lot of my reading was to fill in broad topics along the lines of “what was life like in this period,” not because there was a specific detail I knew I needed, but because I needed a large mass of specific details to draw from in shaping my plot and laying out my scenes. And often one of those elements would suggest a new dimension to the story, so then I’m off down a new fact-reading rabbit hole; rinse and repeat until my deadline starts breathing down my neck and I have to quit adding to the pile.

The other kind of research is one I used to do all the time — but I didn’t really think of it as “research” back then. It was just, y’know, my life. I took an odd assortment of classes and read an odd assortment of books, and they all poured material into my head, and out of that came stories. This is reading for fodder, and I’m finally back to doing it, because I have several projects in the hopper that are all secondary-world, as opposed to urban fantasy (the Wilders series) or historical fantasy (Onyx Court) or what I think of as world-and-a-half (Memoirs of Lady Trent, halfway between historical and invented). It isn’t that I won’t wind up using specific details out of what I read; the difference is that in the end, I’m not actually writing about those things. Lately I’ve been reading a book on Mary Wollstonecraft and Mary Shelley, Everyday Things in Premodern Japan, the Mahabharata, a book on the Sumerians, a bunch of Wikipedia articles on ancient Greek philosophy and society because I finished Jo Walton’s The Just City. Am I planning on writing anything set in Georgian England, Tokugawa Japan, ancient India, ancient Sumer, or ancient Greece? Not necessarily. But it’s all going into the mental compost heap, to intermix and break down and become fertile soil for ideas.

Some subconscious part of myself feels like I’m skiving off of work reading these things, because it’s been trained by nine books of historical or quasi-historical fiction to think the only real research is the kind done for facts. I need to do this, though, or else the worlds I invent will stay firmly in the box of “modified analogues,” places that can easily be mapped to single real-world origins. I need to throw a bunch of different things into my head at once, so that I come up with a society where there’s a deified emperor (a bit Roman, a bit Egyptian) and a caste system (a bit Indian) with a meritocratic way of changing your caste (a bit Chinese) and a clockpunky tech level (a bit Italian Renaissance) and so forth, without it being straightforwardly any of those things. If they wind up having an architecture a little bit like Tokugawa Japan or a schooling system like ancient Sumer, it will be because that happened to click into place, not because I had to use one of those societies for inspiration.

As I said at the beginning, these aren’t clearly divided types. “What was life like in this period” is closer to being a fodder-type question than “how rapidly did the plague take hold in 1665,” because it’s designed to help me come up with ideas for that specific period. And you’ll see the Mayan calendrical system with a minor fictional paint job showing up in Lightning in the Blood because years ago I read about it for fun and wound up incorporating it into a story more or less wholesale, complete with fiddly little details about Year-Bearers. But it helps me to remember that fodder-type reading is a form of research, and one that’s very necessary for my job.

Originally published at Swan Tower. You can comment here or there.

Crossposts: http://swan-tower.livejournal.com/797319.html

There are certain kinds of transition scenes I detest writing. One of them is the “holy shit, the supernatural is real!” scene common to so much urban fantasy; it was a source of great pleasure to me that I could more or less skip that scene in Midnight Never Come, on the grounds that the reaction of a sixteenth-century gentleman would not so much be “there are faeries under London?” as “there are faeries under London?” (You’ll note that nearly every pov character for the remainder of the Onyx Court series already knew about the fae by the time they showed up in the story. This was not deliberate, in the sense of being a thing I consciously decided to do . . . but I wouldn’t call it an accident, either. The sole exception that leaps to mind is Jack Ellin, and I had more than enough going on in the story to divert him, and me, while that transition happened.) It’s boring to me because the audience already knows the supernatural is real (or at the very least has no reason to be surprised by this fact), and we’ve seen that conversation so many times, making it fresh is really difficult. Your main hope is to undermine it in some fashion, like the time on Buffy when they told Oz vampires and demons were real. “I know it’s a lot to take in –” “Actually, that explains a lot.”

I’m dealing with a similar kind of thing in the fifth Memoir right now. The scene isn’t about the supernatural being real; it’s a different kind of transition, one I don’t really have a name for. And of course I can’t get into specifics, but it’s one of those deals where something very complicated is going on, only the complication is of a type that doesn’t actually make for great narrative. After the initial drama of the moment is over, there’s a lot of explaining that needs to happen, and a lot of very tedious suspicion that can’t be laid to rest with the right words or a single decisive action. Inside the story, the whole thing is going to drag on for days — probably for weeks. Making the reader sit through all of that would be dire, starting with the fact that I would have to write all of that.

It’s at moments like these when I love the retrospective, consciously-framed first person viewpoint of this series. Because I can 100% get away with Isabella saying “what followed was very tedious and dragged on for weeks, because there was nothing I could do that would resolve it with a single decisive action. But X, Y, and Z got settled — not without a great deal of wrangling and suspicion, but settled all the same, and now let’s move on to the next interesting bit.” Any viewpoint can skip over things, but this one gives me greater latitude to summarize what I’m skipping, without making it seem like the elided material is simple to deal with in real life. Isabella can acknowledge all the complications without getting bogged down in them.

I had no idea, when I started writing this series, all the advantages that would come with framing the entire thing as a series of memoirs. It just seemed like a period- and subject-appropriate way to approach the whole thing. But my god . . . it’s probably the best craft decision I’ve made all series long.

Originally published at Swan Tower. You can comment here or there.

Crossposts: http://swan-tower.livejournal.com/757823.html

Profile

swan_tower

Swan Tower

April 2025

S	M	T	W	T	F	S
		1	2	3	4	5
6	7	8	9	10	11	12
13	14	15	16	17	18	19
20	21	22	23	24	25	26
27	28	29	30

Syndicate

Page Summary

Style Credit

Base style: Strata by yati
Theme: Altair by dancing_serpent

Expand Cut Tags

No cut tags

Page generated Apr. 23rd, 2025 08:50 am

Swan Tower

Entries tagged with how i do it

Navigation

The not-lost art of eloquence

cui dono lepidum novum libellum

leverage the tricks you have

How to make your CE's job, and your own, easier

picking a series type and laying a foundation

towards some thoughts on series

The Non-Series

The Setting Series

The Cast Series

*The Reset Button Series

The Episodic Growth Series

The Episodic Arc Series

*The Perpetual Motion Series

The Metaplot Series

The Single Book

Novella thoughts

Sequencing

Units of Fiction IV: Attention and Focus (Chapters)

Units of Fiction III: Attention and Focus (Scenes)

Units of Fiction II: Attention and Focus (Paragraphs)

Units of Fiction I: Mechanics and Pacing

The weight of tiny details

If You Ain’t Got That Zing

Ten pounds of story in a five-pound sack

"Where do you get your ideas?"

The Art of Cover Copy

The world's most scattershot progress

Two kinds of research

transitions, blech