Second Person and Present Tense: Why and Why Not

I risk coming off like a grumpy old man in this post, which is something I’ll just have to live with. It’s my assertion, though, that second person viewpoint and the present tense are overused in modern writing, and that outside of certain usages they should be shunned.

Let’s start with second person, that is, when a writer addresses the protagonist of their story as “you” as if they were telling this character their own story. This is something you almost never used to see. I remember an English teacher I had in high school telling me there was no such thing as second person–that it had never actually been done (he was wrong, of course, but it was so rare back then I didn’t know how to contradict him). You see it more and more these days and while I think there is a place for it, it’s almost never used correctly.

If the narrator is describing recent events to someone with amnesia, perhaps, or describing events that have been foretold but have not yet occurred, then second person might be justified. The main and most important use of second person is in interactive stories–choose-your-own-adventures, interactive fiction games, and the like. I used it myself in my experiment to write a novel on Twitter, which allowed readers to pick each plot development by poll.

Otherwise, second person always comes off as affected, as pretentious, and it distances the reader from the writer in a highly artificial way. Which is not to say that’s always bad! Distancing is a valuable technique, for some stories. If you’re going to use it, though, you should have a very good reason–and the fact that it’s trendy, or cool, is not a good reason.

(Just as a tangent here I’ll say I’m not crazy about first person, either–I like limited omniscience in my narrators, and the freedom that provides to expand a story beyond a narrow range of perceptions. But there are plenty of excellent reasons to use first person and it never really bothers me when I pick up a book with a strong protagonist’s voice).

Writing narrative fiction in the present tense isn’t quite as jarring, but I feel it’s getting overused as well and it comes with its own raft of problems. Present tense suggests immediately to the reader that the story hasn’t been finalized, that the events described are still evolving, which means they can’t be predicted–that the reader who is coming along for this ride cannot be guaranteed a coherent or even complete story. It’s a subtle psychological effect and one that needs to be considered carefully.

The writer who employs the past tense when telling a story is making a compact with the reader. It says that the events that are about to unfold, having already happened, can be examined thoughtfully and with a certain authority. Present tense throws that away. Again, there could be good story reasons to do so. Yet drawing on past tense puts your story in a comfortable and established mold that readers have come to accept as the standard for storytelling. It helps speed along immersion and makes the reader feel like they’re in safe hands. You need a good reason to eschew that comfort level, and more often than not I find present tense narratives lacking in justification.

The main explanation for why people use second person or present tense, I am told, is immediacy. The idea is that a story being told directly to the reader–and only the reader–or one told as it is literally happening is better at pulling the reader in, in making them feel like they’re being dragged along on a breathless adventure. I can see the logic in this argument, but I find it rarely works that way. Typically when a writer starts out in present tense, my immediate reaction is to roll my eyes. When they start in the second person I frown and wonder why they made such an odd choice. But even this dubiety doesn’t last. Typically I pay attention to a story’s tense and viewpoint for the first couple of pages–then learn to ignore it, to put it aside and focus on the plot and characters instead. Whatever immediacy the writer has laid claim to disappears as I sink into the work. Writing is always about choices, and when the writer chooses one of these pretentious techniques it only ever puts me off… for a little while. It’s usually not worth it.

It’s possible I’m missing something here, and I’d be happy to hear from other writers who find second person and present tense useful in their writing. But for myself, I’m going to use them sparingly, and only when I can point to an excellent, organic reason for them to be there.

Flatness and Feeling: Three Recent Works

Note: The following post contains minor spoilers for the plots of All Systems Red, Ancillary Justice, and Blade Runner 2049.

Probably the major theme of recent science fiction has been the way technology distances us from our own emotions. One of the devices authors and directors use to explore this distancing effect is intentionally flattening the affect of a central protagonist, exploring the world through their unfeeling eyes to question and problematize our own relationship with the world. Flatness is a tool, and like any tool it can be used to greater or worse effect depending on the choices the author makes. I want to explore three recent works and try to see where they succeed in wielding flatness, and where they fail.

All Systems Red, by Martha Wells, is a novella told from the perspective of Murderbot, a heavily-modified human clone working security for a scientific expedition to an unnamed planet. Murderbot’s emotions have been medically scrubbed, and its main reaction to the world around it is boredom. Even when its clients are endangered and it is forced to protect them, the emotions this creates are awkward and painful to Murderbot, and it acts in ways to escape them. The novella has gotten a lot of hype recently and it’s a nice character sketch but I think it’s the least successful of our three works. Murderbot’s perspective, while compelling, is never really challenged by the story. Almost always, when the humans in the tale act emotionally or with any kind of humanity, they are shown to be foolish and even suicidal for doing so–Murderbot is hardly a Mary Sue, but it does solve every problem in the story through the application of pure logic. Furthermore, anything Murderbot doesn’t care about (which is pretty much everything) is given short shrift here. Early in the story Murderbot fights a giant alien monster. It should be an amazing scene, but it fails–we get Murderbot’s clinical analysis of the creature but no actual description. It can’t even decide if the monster has teeth or cilia. Later on we find out why this planet is special as a setting, and Murderbot dismisses this vital bit of plot information in a couple of sentences. The problem with All Systems Red is that there’s no contrast. We get Murderbot’s grayscale world but nothing else. No visual description at all. The human characters are almost interchangeable and there’s very little interpersonal conflict.

Ancillary Justice by Ann Leckie fares a little better, though I still had problems with it. The main character is the last survivor of a hivemind made of former prisoners of war who have been conditioned to be unfeeling and perfectly loyal soldiers for their own conquerors. The flatness here is leavened–the protagonist is allowed to feel affection for her former officers, and in fact appears to be on a revenge mission (we learn later that it’s much more complicated, and much less sentimental). The action takes place on three different worlds: a marshy planet only seen in flashbacks, a lonely ice world, and a space station with complicated social hierarchies. Leckie does a fair job describing the two planets in compelling detail (the station not as much) and there’s a set piece involving an ice bridge that adds some much needed action to a very cerebral plot. The flatness of the story works against it much of the time, however. Most of the story’s action is described in the briefest possible terms, to make room for long passages of guarded dialogue. The actual plot, which remains mysterious almost until the end of the story, is convoluted and never allowed to evolve organically. We are given mention of aliens and space battles but these are abstracted away, pawns in a five-dimensional chess game where nothing really matters but who wins. The flatness here is a mark of intellectual superiority: the few characters who do show emotion are either mocked or despised for it, while the cold and callous logic of the protagonst and antagonist are celebrated and far more effective. While Murderbot wrestles with its vestiges of humanity, the Ancillary works hard to get rid of hers–she wants very much to be a spaceship again, not a person, and this goal is seen as worthy. This isn’t a failure of the story, mind you. The whole point of Ancillary Justice is that its universe is far too big and impersonal for humanity to run, and it needs to be administrated by beings with greater mental capacity. It’s an interesting theme but one that left this human reader a little cold.

Blade Runner 2049 is the most successful work I want to look at today, one which uses its flatness as a perfect counterpoint to its deep emotional themes. Like most film noir, the movie employs a cold, cynical tone that is betrayed by the deeply human story it wants to tell. Its protagonist, K, is a replicant of a new series, one which is free of human desires. He moves deadpan through a world in crisis, performing a job. A job he has no emotional investment in–to the extent he isn’t even bored by it. He’s been designed from the ground up for flatness and his lack of reaction throughout the film is brilliantly portrayed by Ryan Gosling as both incredibly useful and–to the viewer–emotionally terrifying. All of his relationships are abstracted, through-a-glass-darkly versions of normal human interactions, and his final moments in the film are wonderfully understated. His flatness is wielded here like a blowtorch–he forces the viewer to engage with the things he refuses to touch. His opposite numbers in the film, Luv and Joi (what great names), exist at an even further remove and serve to keep his flatness emotionally grounded. When we see our first real human character in the movie his over-reactions and scenery chewing would almost be funny if they weren’t so heart-breaking, an irruption of feeling the movie seems unable to contain. This is flatness used with precision, by a master.

It’s not entirely fair to compare the two novels to a movie, of course. Blade Runner makes extensive use of its visuals to prop up K’s flatness. The lush colors and surreal set design keep the viewer awake through what could have been a very sleepy first act, definitely, and the star power of the actors compensates a lot for the emotional flatness. Yet I think authors can take a lot of lessons from this movie, all the same. We need to always remember that flatness is a device. Whether we want to praise emotional detachment, like Leckie, or just find it awkward like Wells we need to keep it under control–right up until the moment we need to lose that control and let sentiment overwhelm us.

The Lessons of Creepypasta

First there were campfire stories–which probably date back to the invention of language. These begat “urban legends”: the Hook, the call coming from inside the house. The Russian scientists who accidentally drilled into hell. Stories without authors, folk tales for a more scientific age. Creepypasta is the direct descendant of that canon. It has its own unique features, true. Typically creepypasta does have a listed author. Like all things in the internet age it competes with itself–stories get ratings, get YouTube reviews. Get followers. Creepypasta generates fan art, and derivative works, and even wikis. Yet creepypasta is experienced best the same way these stories ever were: alone in the dark, shared from friend to friend as a kind of rite of passage. Maybe you can’t toast marshmallows with the light from a touchscreen, but you can still scare yourself silly.

As a horror author I’m fascinated by these stories. So often they’re dismally written and threadbare. Sometimes, though, they pack a terrible, visceral punch in such a short word count. A good creepypasta is like horror haiku.

I want to explore some of the things I’ve learned from reading far too much creepypasta. Below I’ll make reference to many of the stories as if you’ve already read them. If you haven’t, they can be easily googled, so I won’t bother with synopses.

Creepypasta is Fast Fiction

The best pastas are short and to the point. They don’t waste time on deep character studies or establishing mood. You’re already a little scared, clicking the link. That’s enough. We get, typically, a nameless narrator setting up the plot, then a scene or two of rising tension, followed by a big nasty reveal. In many ways creepypasta shares its structure with jokes more than short stories: introduction, complication, punchline. Candle Cove is less than a page, in its earliest (and best) versions. Slenderman is mostly just a couple of forum posts. Longer works like the Russian Sleep Experiment still hold out the promise of a final awful epiphany, but as is the rule in all things, the bigger the buildup the bigger the resolution had better be. Longer pastas are routinely downvoted. Fans come for the quick rush of fear–boring your readers is the kiss of death.

Lots of Villains, No Heroes

Jeff the Killer, Slenderman, the Inverted Mickey of Abandoned by Disney–creepypasta loves its monsters. They tend to be visually interesting (it’s easier to make fan art, that way) and often their mere appearance is enough to scare the narrator into running or screaming or having a heart attack. In fact, we rarely see them at work. Their crimes are often second-hand, mere rumors of atrocity. A bloody mouth or a wild look in the eyes is enough. If we never actually see what they do, our imaginations can run wild–a man with a hatchet is scary, a man chopping up body parts is just, in Raymond Carver’s wonderful phrase, “popular mechanics”. Furthermore it might push focus onto the victim, which is a big no-no in creepypasta. While the killers may be richly, even floridly described the victims and especially the narrators are usually cyphers–they almost never have actual names, nor are we ever told what they look like. They exist on the other side of your computer screen, speaking to you through anonymous media. The fact that the killer pushes through into reality is the scariest thing about them.

Stakes are High, or Pointless

If the protagonist actually wants something in a creepypasta, the stakes in a story tend to be ridiculous, to justify potentially suicidal behavior. In one version of No-End House, you win an astonishing amount of money if you can make it through to the final room (you won’t). Some protagonists, like Orpheus, want to find and restore their dead or missing loved ones (it never works). More often than not, though, there are no stakes at all. Protagonists exist in creepypasta for one reason: to regret their own curiosity–which, of course, makes them us, copies their identity onto the reader who similarly was unwise enough to click on a link they knew would scare them. In many pastas the inciting motive of the “hero” is a simple compulsion. They couldn’t not look, they couldn’t not explore the abandoned Disney park, they couldn’t resist opening that letter with no return address or looking at the one image file on the thumb drive. They exist simply as puppets of fate, victims of a universe that actively wishes their demise.

The Universe is Self-Aware, and it Hates You

If Lovecraft dragged horror fiction into the twentieth century, he did it by throwing away the religious and mystical baggage of the nineteenth. God couldn’t save you from Cthulhu and holy water didn’t kill space vampires. The universe is a cold and uncaring place, and you are contemptibly small. Creepypasta, as the horror literature of the twenty-first century, wants to take it a step farther. The cosmos isn’t just a cold void, it’s also a seething mass of disdain and hatred. There is only one way to survive the events of a pasta, and that is to be left alive so you can tell the tale (and even then, you know the evil hasn’t forgotten you, and your time is limited). Many pastas, like Return to Earth, don’t even allow that level of grace–the narrator is telling his story to no one, watching his own demise creep closer, knowing that no one will ever hear what he says. Reality in creepypasta is fluid, malleable, but it only ever bends one way–toward destroying you. No one in a creepypasta ever learned a spell from the Necronomicon to push the nasty things away. And where Lovecraft saw a kind of hope in ignorance–a delusion that brought blissful sleep–in pastaland the horrors seek you out, through message board posts and text messages. There’s no way to escape, or even deny what’s happening.

Style: Distance and Outsiderness

Creepypasta always comes at a remove. The story you’re reading started with someone noticing a strange link on a web page, or they find hidden files on a game cartridge. Candle Cove’s entire substance is just the narrator remembering a strange television show from their youth. The writers of these stories know that you’re bored with real life. They know you think the world is empty of the supernatural, and so they build a firewall between you and the horror–one which always turns out to be more permeable than you thought. Distancing techniques can also add verisimilitude to a story. The great evolution of creepypasta, the SCP wikia, follows a rigidly anti-entertainment format (there’s a reason the containment procedures come first) and an enforced clinical tone to separate you from the horror. Then it pulls the floor out from under you. It’s a great trick. Because pastas are “documents” that you, the reader, have found you can open them safely, but their very artifactual nature makes them seem more realistic, and therefore less escapable. They force you to engage with, and even participate in, the squick. Creepypasta that eschews the clinical tone has its own distancing technique, one based on reader expectation. The best pastas use their outsider status to great effect. This isn’t some polished story that Stephen King sent to an editor for feedback. This is the breathless recounting of a desperate survivor. A madman’s ravings that no publisher would ever touch. The narrator announces at the beginning that no one would ever believe what he has to say–forcing the reader to enter into a devil’s bargain, the one wedding guest who was forced to listen to the Ancient Mariner.

Conclusions for Horror Writers

Creepypasta is very much an anti-style, a kind of radical reimagining of horror that places it outside the more traditional, more established venues for writing. Channel Zero, SyFy’s attempt to turn pastas into television (just like Freaky Links before it, and a dozen other attempts), misses the central point of the creepypasta format–that it’s something you discover in secret, something you weren’t supposed to see. Not something that gets endless promos and celebrity hype on basic cable. Horror writers can’t just lift creepypasta for their own ends–it’s a type of storytelling that only works in short chunks on a computer screen. Yet there are lessons to take away here, messages from a new wave of horror fandom that has its own values and desires. As horror writers, whether we work in novels or short stories, we’d be foolish not to look at the audience reaction to creepypasta and tailor our work to appeal to this new generation of horror fans. We can respond to the cosmic malevolence of creepypasta, eschewing the more rationalistic horrors of the past. We can examine distancing techniques for use in our own work, and rethink the levels of deep characterization we give to our protagonists (rethink, mind you, not necessarily abandon). If nothing else, creepypasta can serve as wonderful inspiration. I’ve read enough horror novels in my life that I’ve gotten pretty jaded. Evil clowns and zombies don’t really scare me anymore. I’ll freely admit that the pictures of Slenderman did, that Abandoned by Disney got my pulse elevated. Creepypasta works as an incredible laboratory examining what still scares us–and what always will.

Why I Don’t Believe in Willful Suspension of Disbelief

We’re told, almost certainly apocryphally, that one of the earliest movies was of a train heading directly toward the camera, and that audiences who saw it ran for their lives. We’re told that a truly great piece of fiction makes the reader forget they’re reading a book (or seeing a movie, etc.) and immerses them entirely in the secondary world of the story.

I’ve never bought it. Oh, I’ve read plenty of immersive stories in my time, but every single time the thought in my head was not “this is real! If the characters die, I will too!” My typical thought is more along the lines of “oh, this is really good! How can I pull off this same effect in my own work?”

Maybe it’s because I’m a writer, and I know how the sausage gets made. Maybe my insider knowledge of how stories are put together keeps me from enjoying them in the intended manner. I don’t think so, though. I think pretty much every consumer of culture knows the basics of how entertainment is made. The water cooler talk about the latest Game of Thrones episode is often as much about the difference between practical and CG effects as it is “can you believe character X did that to character Y?”

Then there are the commenters. In one of my books I suggested that a given handgun could hold thirteen bullets (you get a prize if you can guess which book this was. The prize is that you briefly made me feel like I exist). I got an email from a reader who told me he had enjoyed the book up to that point, but really, since said gun could actually hold fourteen bullets, he could no longer appreciate my work and would not be purchasing any of my future books. I actually took this seriously and mentioned, in the second volume of the series, that the gun could hold fourteen rounds.

I then got an angry email from a reader saying that if I didn’t know the gun could only hold thirteen bullets, I wasn’t fit to be writing about vampires.

I mention this story because I think it shows that we, as media savvy entertainment consumers, have stopped willfully suspending our disbelief–if we ever did. Just as some people can watch a movie and check their texts at the same time (horrible people), when we read a book we’re never completely lost in the imagined world. We’re always–always–comparing the story to what we know, to what we’ve read before. We run to wikipedia or at least the dictionary if we find concepts or even just words that are unfamiliar in our texts. We discuss our stories endlessly–Chris Hardwick has made a career out of dissecting TV shows the very second they’ve finished airing. I don’t think disbelief enters into the equation at any point.

Instead I think we are actively participating in cultural transmission. Rather than passively submerging ourselves in words and images, we are all, in our heads, remixing them continuously into nuanced interpretations, hot takes, and snarky scorn. We don’t just absorb texts, we study them, mining for ideas for our own work, or at the very least for sick burns we can share with co-workers.

As media becomes more and more integrated into our psychological landscape we become more adept at using it, rather than being manipulated by it (current politics notwithstanding). We always experience texts at a remove, now. But I don’t know if this is purely an artifact of the twenty-first century. I’ve always been interested in mythology, and one of the first questions you end up asking when you read Bullfinch is whether the Greeks really believed that the sun was a chariot with Apollo at the reins. The answer of course is that some people did, while others simply used the story to organize their lives–just as many people today seem unable to understand the news or the actions of their friends without framing them in terms of scenes from Harry Potter. I think this may be something intrinsic in the way we tell stories.

Think of a stage magician’s show. We are told that the magician will saw a woman in half. We watch it happen. Yet only a very small part of our lizard brains actually thinks it’s real. The vast majority of our consciousness is instead devoted to trying to figure out how the trick is done. We engage with the act on multiple levels at once, almost all of them on a metatextual level–we are outside the thing we perceive, aware of factors beyond what the magician wants us to see. The same can be said for every book we read, every television show we watch. Really interesting art often plays with our expectations and our preconceptions. Think of unreliable narrators, antiheroes and stories that break the fourth wall.

When Coleridge coined the term “willful suspension of disbelief” in 1817, he was specifically talking about genre stuff (what he would call “fantastical tales”) being leavened with verisimilitude and human pathos to make it more palatable. His idea was that readers would scoff at fantasy stories unless they were given enough realism to have something to hang onto. I’m not sure if this was true even at the time; I’m certain it isn’t, now.

What’s the takeaway here, for writers? Be conscious, always, of the fact that your reader isn’t reading your work in a vacuum. She has access to endless secondary sources. She can call you on your bullshit, or she can get in a discussion group with hundreds of other people about the details of your fiction. Don’t let that paralyze you, though. Instead, play with it. Withhold details or intentionally provide false information–which you correct later, for dramatic effect. Don’t be afraid of direct address or other techniques which might damage the illusion of secondary reality. But always remember, when you write, you are not alone. You are not screaming your truth into a void. You are contributing one long twitter thread in the culture conversation that began before you were born and will long outlive you.

Was that helpful? Or are you already rolling your eyes, thinking about how I overwrote that last sentence, and how much of this article I got wrong?

Dream Logic: Using Surrealism in Fiction

I heard a podcast recently where the guest talked about the Magical Realist novel he was working on. The host asked, “is that just a fantasy novel but you don’t want to admit you like geek stuff?” The guest laughed and admitted that was pretty much accurate.

Similarly, you hear a lot of people go on and on about “dream logic” as if it’s an excuse not to have to write a plot for your story, or have characters with realistic motivations.

Neither of these things is remotely true. Oh, it’s true that some writers use these terms as excuses. But they shouldn’t.

Surrealism is a style, not an excuse. It can be incredibly powerful when it’s done right. You look at a Salvador Dali painting of people wearing lobsters on their heads and your first reaction is to laugh. Then you move away quickly, and the image haunts you every time you think about it. That’s not just Dali throwing weird crap on a canvas and calling it art. The whole point of Surrealism, according to its creators, was to tap into the subconscious. To make connections between things that seemed separate. It’s an incredibly hard thing to do, and almost impossible to do well. When it does work, it’s magic.

“Every dream has its navel,” as Freud almost assuredly didn’t say. I’ve always interpreted this as meaning that any story, no matter how far-fetched, weird, or seemingly dissonant, must contain a connection to reality. At the very least some kind of nod to real, lived experience. Surrealism exists on that thread-like connection, the region where the wild nonsense of chaos bleeds into the stark cold light of day. Great practitioners understand that you have to walk that tightrope to bring back phenomenal stories.

When people talk about Dream Logic they always forget the “logic” part. Or the “realism” in Magical Realism. Let’s look at some of these styles individually, and see where, while they promise chaos and insanity, are actually art at its most carefully controlled:

Dream Logic: Often called “Fairy Tale Logic” as well. In some ways, DL is the opposite of surrealism. It’s the attempt to impose consistent, logical rules onto a nonsensical world, and more often than not, the attempt fails (intentionally). If surrealism is about digging in the deep soil of the Id, DL is about the Superego trying to make sense of a messy room. Alice in Wonderland is dream logic at its finest–Alice is a scientist attempting to solve Wonderland with clear syllogistic logic. The story is about what happens when you push too hard and the crazy thing pops back up behind you. David Lynch is a modern master of DL–Twin Peaks: The Return is a master class in the style. Be careful, though. Readers who expect straightforward narrative will quickly get bored with this “crazy” style. You need to find a way to keep them turning pages.

Stream of Consciousness: Nicholson Baker’s Mezzanine is one of my favorite books. It’s the story of a man riding down an escalator, letting his thoughts wander. James Joyce’s Ulysses is often held up as the one and only masterpiece of SOC. It’s not, though. The real master here is Stephen King. King gives some terrible writing advice in his book On Writing: create an incredibly realized character and then just see what happens to them. No one should ever do this. King gets away with it because he has a genius-level aptitude for creating stories out of random events–his brain just naturally gravitates toward a loose, swingy plot, so eventually he gets a book out of his wool-gathering. Do you have Stephen King’s instincts? No? Best to use this for, say, a chapter in a longer work, or just in flashbacks, unless you’re absolutely certain you can bring it back home.

Magical Realism: Argh! No, it’s not just fantasy! It’s a reaction against fantasy! Magical Realism is an attempt to describe the real world–the place you’re living right now–period. The conceit is that the only way to truly explain the beauty and importance of a real place, person, or event is to highlight it with fantastic symbols. Some life events are so big and so impossible to describe that we need metaphors to handle them. The loss of a loved one becomes a story about them growing wings and flying away. But the story, the damned story, is about the loss, not the wings. This might be one of the most misused terms in literary history.

Dadaism: Not just surrealism’s cranky punk rock grandpa. The two movements actually serve radically different ends. If surrealism mines for hidden connections between disparate subjects, Dada is all about breaking apart the established connections of our every day life. A toilet seat covered in velvet is surreal; one covered in spikes is Dada. It forces you to rethink the things you do automatically all day long. A metronome is something you watch. Unless there’s an eye on it, then it’s watching you, pal. A man puts a bicycle wheel on a stool and puts it in a museum. Not just to prove that art is bullshit. No! That was the opposite of the point! The point was to show that objects you encounter everyday become invisible to you. By recontextualizing them they become visible again, which is one of the most jarring experiences you can have. Dada was about iconoclasm, about smashing your face into consciousness.

The last thing to say about invoking surrealism in your work is that it’s dangerous to the writer, first, and the reader, second. The writer who dallies with the unconscious is liable to end up discovering things about themselves–their own personal internal symbol systems, phobias they didn’t know they had. If you find that happening while you’re writing, it’s actually a sign that you’re on the right track. You owe it to yourself to explore those strange feelings. To dig deep and see what you can find. Write crazily. Try automatic writing, where you just type without looking at the screen. Throw a random symbol in your work and see what shakes loose. Just remember–you have to edit it all, later. You have to make it make sense, make sure the reader never feels like you’ve lost control. Whatever happened while you were composing is yours to keep, and doesn’t have to show up on the page.