I have a better understanding of how to write books than how to revise them

revisingI’ve written first drafts of lots of books, but I’ve only ever brought two books from first draft to submittable form. And in both cases, it was a bit of a haphazard process. With This Beautiful Fever, I did real editing. In fact, I deleted the whole first third of the book and rewrote it. But with Enter Title Here, I did surprisingly little. Most of what I did was cutting stuff. In the end, I cut about a third of the book. Which is hard, of course, but it’s also a very compartmentalized process. I’d wake up and go through it page by page, asking whether this scene, paragraph, sentence, or word really needed to stay.

Now I’m trying to revise my sociopathic mom book, and it’s proving a bit less tractable. This book is a big one, for one thing. It’s 110,000 words and it spans three years and lots of different events. The book is a very complex machine.

Right now, the thing I am wrestling with is the character’s emotional and intellectual development. It’s astonishing how you can write an entire book and yet not be entirely clear about the character’s journey or whether they change. In this case, there’s just something about the mother’s journey that’s not quite sharp enough. She never quite comes to terms with her own behavior. In the end, it’s not even clear if she understands it.

I think I’ve figured out a solution (I’m going to weave on extra thread into the narrative). But it’s a bit exhausting to think about going through the whole manuscript and patting everything into place.

And then after that there’s everything else: the cutting of extraneous words, the revising of awkward sentences, the checks for internal consistency. It’s all such a big production. And I think it’s going to take at least a month.

In the case of at least three novels, I’ve gotten to this point–the place where I’d need to spend serious time polishing the novel–and decided that the underlying product wasn’t strong enough to warrant the effort. In this case, I don’t think that’ll happen…but that still leaves me doing all this work. Sigh.

Up to my 49th draft of chapter one of this children’s book

rewritingI thought that the 48th was going to be it (I got all the way to Chapter Four!) but then it turned out not to be it. All of this rewriting seems crazy now, but, in my opinion (and in my experience), a book’s not gonna work unless the first chapter works. For me, the first chapter contains the DNA of the entire book: the characters, the setting, the milieu, the conflicts, and, most importantly, the voice. You can’t go back into a novel and add in the voice. Unless you’ve got it crisp and clear in the first chapter, the rest of the book’s going to be limp. When I was writing the first draft of my sociopathic mom book (which has a very distinctive voice), the voice was literally the last thing that gelled. I’d written 70,000 words (in various drafts) and was back at page 1, trying to make it work and wondering what was missing. And then I had a key realization regarding the mom’s motivations, and I wrote a 900 word scene and it was all there. I remember re-reading that scene ten times before I went to sleep that night and reassuring myself that I’d finally found it.

After that point, I wrote the rest of the book (all 110,000 words) in fifteen days.

Before I can move on with the rest of this book, something like that has got to happen.

The way that I prevent myself from going crazy is that every night, I develop a new strategy for tomorrow. That way, there’s always hope. This time, I’m actually pretty excited. I’ve got a whole new tack.

But only time will tell…

As opposed to the last few times that I’ve talked about this, I’ve now actually sold a novel, so (hopefully) people won’t chime in with condescending comments about how I just need to turn off my internal editor.

Believe me, I do not have a very overactive internal editor. My internal editor is the same as the thing that makes me want to write. If I’m interested and compelled by something, then I have no problem writing onwards, despite its flaws. But if I’m not, then it’s a warning (to myself) that I need to step back and reevaluate.

You’ll see! You’ll all see!

Sometimes it feels like real life is something that only happens behind a keyboard

Still coming down off the process of writing the first draft of Mommy Has Empty Eyes. Writing that draft has to stand out as the most intense four weeks of my life. Even thinking about it makes me shiver. I’m always pretty fast with my first drafts, but that’s the first that I’ve subsumed everything to the writing process, and that’s the first time that I allowed myself to explore my doubts, rather than ignoring them. I don’t know whether the result is good. Well, I know it’s good, but who knows if you’d agree. The point, though, is that it felt really good. It felt right. And coming back to real life has, in some ways, felt like a bit of a letdown. It reminded me of a passage from this essay (which is about how writing is primarily something that you do alone in a room, and that your success as a writer will depend upon how you bear up under that solitude).

The room can become a hole. Your talent of the room, your ability to be there with all your soul, can overwhelm you. Then the rest of life becomes unreal and, worse than unreal, a kind of unlife. So you find yourself writing with a very sophisticated consciousness but living in your relationships with other people far beneath what you write, because it’s gotten so you only really exist in that room and you don’t care about outside. And since you write necessarily from memory — for writing in a sense is memory, is what you cared about yesterday, or last month, or in your childhood — your lack of feeling for the present may not show up in your work for a while. But when it does, you’re through. You may still be published, still make money, still be read, but people won’t care the way they used to — and they’ll know it, and they’ll let you know it.

You would think that _knowing_ you’re going to feel bad would somehow inoculate you against feeling bad

th_BadStoryDroWhen I was in the final stages of writing the first draft of the novel, I was feeling so good that I knew the feeling couldn’t last. When you’re in a good mood, it’s impossible to fathom a bad mood coming. But I knew it was on its way! I had no idea what I was going to feel unhappy about, because there’s really nothing in my life that is particularly bad, but I knew it was coming.

And here it is.

Now that the semester has started, I need to write some short stories for workshop, and it is proving to be extremely slow going. Last semester, I was on fire, producing much more than I needed to for workshop. This semester, not so much. The world simply seems drained of meaning. I can imagine all kinds of stories, but they don’t seem interesting. It’s hard for me to imagine what a person could care about or why they would bother.

It’s not a depressed mood, so much as a mild anhedonia.

Still, when I was in this mood last year, I produced at least one extremely good story, so we’ll see. There are only two ways to write when you’re in this mood: the first is to go deeper inside and try to pin down the thing that you need right at this moment; and the second is to get further from yourself and immerse yourself in the richness and specificity of the world.

I am alternating between doing the two.

Writing requires a lot of faith. When you’re working on a story, you need to have faith that it’ll come together (even though there’s a very good chance that it won’t). And when you’re between stories, you need to have faith that something will come along. Without that faith, you end up spending way too much time chasing down every mirage because the act of writing a bad story so much like the act of writing a good one that, as long as you don’t allow yourself to slow down and think, you can very easily convince yourself that you’re doing the latter instead of the former.

Wrote a crime novel!

Let’s not closely examine why I began another novel so soon after completing the last one. I just wanted to, okay! The prospect of working to revise the last one (which was a YA novel) wasn’t very exciting, since it could be aeons before it’s ready to go on submission. But I also didn’t want to waste my seven weeks of uninterrupted free time by working on short stories (it’s hard to get any long-form work done during the semester). Also, it feels really good, physically and emotionally, when you’re in the zone on a novel. and I liked that feeling.

So I wrote a novel. It takes place in a twenty-minutes-into-the-future world where doctors use brain-scans to ferret out psychopaths, and then the police book them on minor violations as a pretext for sticking them in hospitals for the rest of their lives. In this world, a working-class mom schemes to get her daughter into a high-class Northern California prep school that’s designed to turn children into geniuses. However, after the mom tests positive for sociopathy,  she pulls a switch, convinces the police that the scan actually belongs to her daughter (who, in reality, has about as much empathy as one can reasonably expect out of a teenager), and then everyone becomes convinced that the genius school might be able to turn the daughter into a productive member of society. But the daughter refuses to go along with the deception, which means that the mom has to use all her sociopathic cunning to somehow force her thirteen year old daughter to toe the line.

Although I intend to market it as a literary novel, it borrows the typical noir novel structure. Ever since I started reading crime novels a few years back, I’ve really liked the structure of noir literature. It always starts with an ordinary person who’s placed in an untenable situation where something they want is juuuuust out of reach. And then they commit one illegal-but-thoroughly-understandable action (telling one lie, forging one signature, etc) to get that thing. And that succeeds…for awhile. But then they’re forced to engage in increasingly awful behavior in order to conceal their initial crime.

And the book is also about parenting! Something I know zero about. But I’ve watched a lot of family sit-coms, and it’s pretty much just like that, right?

I am very happy with the book. This is the first time I’ve finished an adult-market novel without having the vague suspicion that I was going to abandon the book and never look at it again. I am pretty sure that I will someday revise this and send it out (probably over the summer).

However, it was a brute! I’ve never had to work so hard at dragging a novel, kicking and screaming, out into the world. Remember how at the end of last year, I wrote a post on how long it took to write, edit, and revise one of my novels? I believe the total was something like 160-170 hours (which is absurd, I know. Probably I’ll someday look back and be embarrassed by the smallness of that number). Anyway, this one has already taken 160 hours. My first draft ended up being 110,000 words, but I also discarded about 60,000 words along the way. I restarted the book two times. The first was after I’d written 11,000 words. The second was after I’d written 30,000 words. That sucked.

It was a very complex protagonist, and the book both has a lot of threads and is very plot-driven, and, since the narrator is very unreliable (she’s not lying to you…she’s just very unintelligent, so she doesn’t understand much of what’s going on), I had to know, at all times, exactly what: a) the narrator thought was happening; and b) was really happening.

I also had typical science-fiction novel problems (even though this is not really an SF novel), in that I needed to figure out the structure of things, how the world works, how the genius school works, what people believe about psychopathy, what’s true about it, what the mom thinks, etc.

The result is, though, extremely tight, I think. It fits together like nothing else I’ve ever written. We’ll see, though.

Anyway, I am really excited. It’s a good start to the year. On a sidenote, I keep track of writing stats on a year-by-year basis, and the my stats for this year are insane. I’ve written an average of 7,100 words and 6.2 hours every day. If I keep up this pace, I’ll write 2.5 million words this year =]

“Writing is something you do alone in a room”

_icon83This novel has finally reached the point where it’s flowing really well. I’m about 73,000 words in and can state with a fair bit of confidence that I am going to finish. Even if the semester ends and swamps me with work, I will finish. Even if I get super depressed (which sometimes happens around mid-winter), then I will still finish. That’s not really a matter of my indomitable will or anything. It’s just simple human psychology. If a human being gets 75% of the way through anything, whether it’s a movie or a book or a hike, then they’re probably not going to quit before it’s finished.

But there are still a lot of hours between me and the end. And looking at those hours makes me think about the hours that I’ve already put into it.

Over the past month, I’ve worked in a more thoughtful and more sustained fashion than I ever have before in my life. And it’s been really fun. But it’s also been really lonely. That was my winter break. I didn’t do anything. I didn’t see anyone. I didn’t go anywhere.

It’s not that there’s any specific person or place or event that I regret missing. It’s that all the things I gave up were light time: things that would’ve passed the time in a pleasant way and left behind warm and fuzzy impressions in my memory. In contrast, spending 123 hours at a desk (and counting!) is heavy time. Those hours sit with you very heavily. You can’t escape from them: you have to exist in each and every one of them…which tends to make you very aware of the passing of time.

And living on a different time-scale puts you out of sync with other people. In that way, it’s similar to being drunk all the time. When you’re drunk all the time, life is exciting in a way that it’s not for sober people. Not because you’re doing anything particularly exciting, but because your emotional life is so turbulent. You can go from wanting to kill yourself in the morning to dancing ecstatically at 3 PM to seeing the oneness of the universe at midnight. Because of that, the days separate out, and each one becomes very distinct, and you do crazy and impulsive things because the tomorrow seems soooooo far away. When you live that way, you can’t take very much of it with you–being drunk all the time doesn’t lend itself to accumulating skills, friends, knowledge, or professional success–but the experience of drunken life feels so much fuller than the experience of sober life. The point of drinking isn’t to anesthetize yourself. Or, at least, it wasn’t for me. No, it’s the opposite. The point of most things in life–the consumption of media, in particular–is to make time pass smoothly and easily. The point of drinking is to break up that orderly flow and smash time into little pieces.

Engaging in a project like this is a bit like that too. That heaviness is exactly why it feels like such a worthwhile experience: it makes me feel like I am actually using my time in some manner, instead of just trying to get through it. But the heaviness is also a burden. Because even when heavy time is pleasant, there’s just so much of it. I don’t even know how else to explain it, because it’s totally independent of what you actually do during that time. Even during the happy parts of heavy time, the slowness of the time is quite unpleasant. If someone was to ask us, we’d say that we want life to be heavy (i.e. meaningful and full of experience), but if we were to look inside ourselves, I think we’d find that most of the time, we actually just want life to be over.

I wrote the other day about how I’m only rarely afraid of death. That’s because so much of my behavior (of any person’s behavior, really) seems to imply that I hold my life very cheaply. How can watching fourteen straight hours of television (even excellent television) be interpreted in any other manner than as a desire to stop existing?

*The name of this post is drawn from this essay. I was going to talk about it, but it somehow never came up…

An age-old novel-writing question: How can I sustain a plausible love story using the minimal number of actual scenes between the two principles?

Your handsomeness is not going to fool me into forgetting that you're barely in this book!
Your handsomeness is not going to fool me into forgetting that you’re barely in this book!

I am awful at love stories. A friend of mine was telling me the other day that all her stories are about the guys she’ve dated. I can’t even imagine doing that. It’s not that I dislike love plots. I like them alot. I find them very warm and wonderful. But I just don’t have much to say about love. However, love is a real part of life. And sometimes a novel simply demands a love story. Like, you can’t ignore it. Love and sex often intrude on life.

Which means that I not infrequently am forced to ponder: Can I make this love story work even though there are only maybe five scenes between the characters?

It definitely can be done. Look at Jane Austen. Sometimes her novels have lots of meetings between love interest and protagonist. For instance, Emma and her love interest meet on a number of occasions. But how many pages do Elizabeth Bennet and Mr. Darcy really spend together? He’s very much in the background through much of the book. And yet their love feels plausible. It almost seems like the only kind of love that could be shared between two such formal and terse characters. It’s a love that’s revealed not through words, but through actions.

And, in general, there’s a larger question here about economy of incidents. The easiest way to build intensity and show movement is to have something happen many, many times. If two people meet and exchange numbers and then go on a first date and on a second one and a third one and then it’s their one-month anniversary and etc. etc. etc., then obviously the audience will believe they’re in love. I mean, they’ll be bored, but they’ll believe it. Whereas if you sweep over all that time with a blistering narration, then it’ll move much faster, but might not be as believable.

In general, I have become a much bigger believer in doing everything as few times as possible. I don’t like to look at a novel and see, “Oh and here they have lunch again and here they discuss the previous night and here…” No. I want to be able to sit down and enumerate, in a specific way, every interaction that two characters have. For instance, I want to be able to say: “These characters have six scenes together. In the first, she kicks him in the kneecap because she thinks he’s an Islamic terrorist. In the second, she apologizes to him in the hospital. In the third, he testifies against her in the course of his civil suit for all the emotional distress and physical pain she inflicted on him…”

I think that if you break down lots of good novels (and I’m actually just making an assertion here, since I haven’t done this), you’ll see that in lots of them, there’s considerable economy of scene. Each incident is very sharp and very specific and shows measurable movement in their relationship since the previous incident.

Romantic subplots are just an example here. It’s also true for anything in the novel that requires movement: a job, a friendship, a question to become the world’s most powerful sorcerer…

I write my novels in Scrivener (I know, feel free to groan right now), which allows you to tag each scene with keywords. Lately, I’ve taken to tagging all my scenes with keywords relating to the various plots that are advanced therein. Obviously, the main plot is threaded through (almost) every scene, so there’s no point in tagging that. But all the minor ones, the friendships, the love stories, the sad and lonely declines, get their own tag. I’m hoping that when I’m done, I’ll be able to just click each tag and immediately be able to break out each subplot and see the places where it’s advanced in the novel. Hopefully, this will allow me to learn something about economy of incident. It also might just be a waste of time and an easy way to procrastinate (today I spent half an hour figuring out how to make Scrivener reset the chapter count for each new part of the novel. Yes…that wasn’t procrastination at all).

How writing a novel is like being that one (semi-useless) superhero who can see the future

Novel-writing continues apace. It’s pretty much all I can think about, so I am going to blog about it as well. This is definitely one of the knottiest ones I’ve ever wrestled with. I got to exactly the same place (30,000 words in) where the novel failed last time. And I was about to write exactly the same chapter (like, same general place, conflict, etc) where it failed last time. And I started writing. And I was writing and writing. And I had it all plotted out in my head. And…it wasn’t working.

Man, remember how bad this show was?

So I went down to the kitchen and ate some Wheat Thins and tried to figure out what was missing. And I realized, “Aha, I need to raise the stakes! Shit hasn’t gotten more real in a long time!”

So I went back and made the character start running out of money.

And then I wrote a few thousand more words. Nope. Still having that reluctance. Still having that weird unevenness in the writing and that overreliance on dialogue. This was a bit shocking, considering I’d been swimming along pretty good and had just written one of the best scenes I’ve ever written in my life.

So I lay on my bed and was all like, “Well…dammit. What do I do now?”

And then I realized that everything I’d just written (that first thirty thousand words) could be part one! And then I could just skip ahead three months (over all the boring low-stakes stuff I’d been fiddling around with) and then skip right to to the place where the stakes get higher! It’s pretty amazing. The girl gets into school in one chapter, and then in the very next chapter she gets to be on the verge of failing out. Very exciting moment! I had to spend fifteen minutes figuring out how to reconfigure my Scrivener document as what it calls a A Novel (With Parts).

And during that fifteen minutes, I had another brainstorm. Now that I’d time-skipped once, I could do it again! So I went into my outline and added a Part Three, where I get to skip ahead three whole years! And just for fun, I’m also gonna put in a fourth time-skip (right before the Denouemont).

Oh, now I remember the main point of this blog post. You know how whenever there’s a guy in a movie or TV show who can see the future, there’ll always be one point in time that’s fuzzy–beyond it he can see nothing clearly? And that’s the part where the big climactic event happens, the point where the world could either be saved or lost?

Well, writing a novel is exactly like that (for me). Generally, I can see ahead pretty flawlessly. Like, not just on a chapter-by-chapter or scene-by-scene basis…I can usually even see what the individual beats are going to be in a scene. I mean, they’ll change (sometimes drastically) as I write towards them, but they’re usually there.

And then sometimes I can’t. And it’s not like the power fades away and gradually becomes fuzzier. It’s like, beyond a certain point, everything is vague. Forget about beats, I can no longer even see scenes. And I’ve noticed that this always happens right after a point where there’s a decision that I haven’t yet made. Oftentimes, it’s a decision that I never knew I needed to make, until I’m suddenly confronted by it

But once I make that decision, suddenly entire chapters fall into place in this cascade of images. It’s pretty cool. I mean, it’s also possible that it’s pretty amateurish and that I should be thinking way harder about what to write (just covering my ass in case I later turn my back on this post), but at the very least it’s a pretty exciting thing to have happen.

Nowadays, I’ve gotten much better at seeing the points of vagueness and preemptively answering unanswered questions (I don’t mean questions that aren’t answered for the reader. Any good novel contains many such ambiguities. I mean questions that I, the author, don’t know the answer to, even though I should). But sometimes they sneak up on you. And sometimes you see them up ahead and you can’t think of anything, so you just write towards them and hope that the solution will make itself clear.

Just remember, sometimes the solution is to write a novel in parts.

Incidentally, today I wrote for 641 minutes (10.7 hours). This is my longest writing day (by almost an hour) in the 650 days that I’ve been recording daily writing time. I also wrote 10,250 words, which is my fourth-highest-ever wordcount. The only two days that were higher were the 10 hour frenzy in which I wrote the last 14,000 words of Enter Title Here and two of the final days when I was writing This Beautiful Fever. Since I am nowhere near the ending of this one, I certainly consider this to be a pretty major accomplishment. Although, I guess I did have to cut a substantial fraction of the words that I wrote today…


In this case,

Novels are insanely complicated

This was one of the top image results for "sharpness"
This was one of the top image results for “sharpness”

Got thirty thousand words into the novel before realizing that I had failed to rigorously imagine one key thing (how the mother felt about something that her daughter was doing). And when I belatedly started trying to do some of that imagination work, the entire novel fell apart. Like, just completely unravelled. I realized that the novel had exactly the problem that I described in American Hustle: the characters weren’t emotionally invested in the thing that was taking up most of the screentime, so I’d had to shoehorn in a sideplot in order to give them some life.

Reimagining the character’s motivations meant reimagining her history, which meant changing the entire narrative voice of the story (which I actually hadn’t been too happy with before anyway).

It’s all connected. When one thing is wrong, it wrongs up a whole bunch of other things. In this case, I’d been worrying all through the novel that it was too dialogue-heavy. Even when I glanced at the pages they didn’t feel right: there wasn’t enough variation in paragraph size. I’d tried to fix it up by going back through and adding some descriptive details, but that looked, felt, and sounded like a jury-rigged fix.

I think, in the end, all of that shakiness was a result of my own uncertainty. I didn’t understand my novel enough to be able to know what thoughts and images and actions needed to bubble up at each moment. And I sensed that, so I left it blank, and filled up the pages with talking (which can often be a great cover for a lack of substance, since conversation passes time and feels a bit like action). I mean, I’m caricaturing those thirty thousand words a bit. Lots of interesting things happened, including many things that’re going to remain in the next draft of the novel. But, in the end, they weren’t right.

After going back and reimagining that motivation issue, I was able to write a thousand words and then stopped short. The main character’s daughter was eating ice cream, and I wasn’t sure whether or not the mom was going to scold her about it. And so I went and lay on my bed and I realized that this unsureness was tied to a whole host of other things that I was unsure about (primarily, I hadn’t yet thought about the actual mechanics of the fancy-schmancy childcare center that’s one of the centerpieces of the book). And then that required a lot of rigorous thinking too.

Anyway, I think I got that sorted out, too. As I mentioned before, I normally know that I have the right answer when I sense things becoming more specific. The right answer turns a hand-wavy notation in my outline (something like “He goes on a journey and finds himself”) into “He goes back to his old college and sees that the professor who tormented him is now extremely wealthy because he was an early-stage investor in Google and learns that there is no karmic justice in the world.”

There’s also a certain sort of elegance to the right solution. It’s hard to explain. But wrong solutions feel wrong. They contain too much doubling back and too many tortured motivations. Often I’m able to justify the most insane things to myself by saying something like, “Well, this character acts this one way in this one scene and this other way in this other scene because he’s complex! People change their mind! They act with different motivations at different times!”

Which, yeah, sounds very high and mighty and artistic, but, on an aesthetic level, it just doesn’t work. Fiction should (in my opinion) have a sharpness to it. Because none of this stuff is real, characters, settings, places, situations, need to leap off the page if the audience is going to be able to see it at all. Even when you’re writing a complex, multi-faced character, then you still don’t get to be fuzzy–you just need to make sure that all their facets are sharp.

Novels are incredibly complex. When they’re clicking, you don’t think about that. But every novel–even a realist one–is a whole world with its own rules and its own logic. And you can’t develop that logic simply by deciding “This is how things are in this world.” Writing a novel isn’t like ordering a sandwich at Subway: you don’t mix and match from a menu of discrete elements.

No, you choose each element so that it supports every other element, both on a surface level (so that the plot, character arc, and conflict are sharp) and on a thematic level. And when everything fits together like this, that’s not the mark of a great novel. No! Plenty of bad novels have the kind of cohesion that I’m struggling to reach. Actually, this sharpness is just the bare minimum thing that you need if you’re going to write something that succeeds on any level.

Most novels that I read are so competently-constructed–even the bad ones–that I forget how hard it is to achieve that kind of competence.

And I still don’t know whether this novel is eventually going to come together. I give it about a 50% chance at this point. On one level, it’s extremely disheartening when you work on something and it’s not coming together. But it’s also a bit exhilarating. Generally speaking, I walk around all day listening to a constant internal monologue that’s pretty self-important and banal. And I always agonize about that. I mean, I shouldn’t be wasting valuable brain-time worrying about a rejection or wondering what I’m going to order for dinner. No, I should be pondering important matters.

But when I’m figuring out a novel, my brain works so much harder than it ordinarily does. I can actually feel it turning things around and crunching them into place and taking up hypotheses and discarding hypotheses and framing questions to itself. It’s the kind of thinking that I, when I was young, used to assume would fill up my entire adult life.

When to listen to your fear and when to ignore it

Keep_Calm_Big_ThinkSomething about writing novels brings out fear in the way that no other writing task seems to. They’re so big and there are so many variables and so many decisions. And the price of failure is so high. When you walk away from the wreck of a novel, you lose something–a sense of your own invulnerability–and start wondering, “How could I have worked, for so long, on something that was so bad.”

When I talk about failed novels, I’m not talking about ones that were merely not very good. I’ve written those too. A not-very-good novel still has something to it: a story, some characters, an arc, and maybe a spark of something new. Just, for whatever reason, it’s not very good. But I’ve also written failed novels. And that is the worst. I’ve now written three novels that were so abysmal that I found it difficult to reread them. Once they were done and finished, the thing I realized about these books is that there simply wasn’t any story: they were ninety-five thousand words of smoke and mirrors. Events happened, but they didn’t add up. The books didn’t have that emotional core.

Lots of published books and even more published short stories have this problem. There are three main ways that a work of fiction can fail to be a story. The first is that it can be trivial. If there’s nothing in the work that matters or really resonates with the reader, then it doesn’t matter what happens. The second is that it can be rote: if the novel never deviates from what’s expected of it and offers nothing new, then it barely exists–it’s merely a shadowy something that’s cobbled together out of bits of what’s come before. And the third (which is the problem that I usually grappel with) is that it can be incoherent. If the story doesn’t know what it’s about, then the selection of elements will be governed by whim instead of by its own internal logic. The different parts of the story will work at counter-purposes to each other and dilute or destroy any possible effect (two recent movies that suffered from incoherency: American Hustle and The Wolf Of Wall Street).

In an incoherent book, the failure is a lack of vision. Elements are thrown in because they’re “cool” or because they’re simply the sort of thing that happens in books.

I have never written a book that didn’t start off as an incoherent mess. And it’s always the same. I’ll start writing the book, and I’ll get five or ten or fifteen thousand words in and then I’ll suddenly be terrified. I simply won’t want to write another word. The problem is not that I don’t know what’s coming next. I’ll know. I’ll have a plan. I can tell a person exactly what the next scene will be. But I simply won’t want to write it.

In situations like this, the common advice is to face down your fear and press onwards. And I’ve done that: I’ve ignored that feeling and gone forward and written that scene. And then the next scene. And the scene after that.

In one other case–my first YA book, This Beautiful Fever–the ship, somewhat miraculously, managed to right itself. The result was certainly incoherent, but at least the narrator’s story was fairly clear. A huge number of subsequent revisions managed to at least partially clear up the incoherent elements. And, in the end, it came out as something fairly readable.

But in those three cases (and in at least two other novels that I never completed), I never found my bearings. I just piled scene on top of scene until I’d finally written so many that I felt like the novel could end.

Since the last of these failures (which occurred just this last summer), I’ve learned to listen to my fear. I’ve learned to draw back and say, “Why am I afraid to write this next scene? What is missing here?” And I literally write down lists of questions for myself. And I spend hours in bed, staring at the ceiling, trying to figure out what it is that I’m not seeing. This isn’t about finding the answer. It’s about finding the right question. And when I do, the question always reveals some fundamental problem with the central narrative of the story. For instance: “The way I’ve written her, would this character even care about being a good mother?” or “If God is really talking to her, then why is she questioning it?”

When I find the right question, the temptation is always to gloss it over with some irrelevant bullshit like, “Oh, she needs to pretend to be a good mother in order to stay with her boyfriend” or “She’s still not really sure if the voice actually is God.” That’s all stuff that sounds fine on paper. I mean, you can tell it to people and they’ll nod their heads and say, “Oh, that makes sense.” But writing a novel is not an exercise in bullshitting. It’s not about finding a plausible answer; it’s about finding the right answer.

The problem with the bullshit answers is that they cripple the emotional heart of the story. If she’s not sure that the voice is really God, then the story becomes a weird detective story, where she’s trying to figure out who’s talking to her. And then, when she finally does, then she’s still subject to the same core problem! And then, because there’s no emotion in the main story, I need to insert some through side-story: some drama with her friends or whatever. And the result of the whole thing is a whole lot of flash and glitz, but not a lot of movement. This is exactly the problem with American Hustle. Christian Bale and Amy Adams aren’t really at all invested in the scam that they’re being forced to perpetrate, so the writers had to force the love triangle to bear all the emotional weight of the story.

Luckily, the bullshit answer doesn’t make the fear go away. Whenever I come up with one, I’ll go back to the story and try to write it and will feel like I’m tangling up everything in knots. And I still won’t want to write further.

Trying to come up with the right answer is very frustrating. Sometimes it ends up being an easy fix (“Just excise my chatty voice of God, and turn it into a single mysterious oracular command”) and sometimes it ends up being really hard (“The character I’ve written is not one that can carry the weight of a novel like this; I need to delete everything and then write her in a way that’s fundamentally different”).

In general, I can tell that I’ve come up with a good answer when the answer is something that makes my novel less complicated and more specific. When you’re operating off a bad conception of the novel, then lots of things about it remain stubbornly vague (“Oh, and at this point she has an argument with God”). But when you have the right answer, those things pop into focus (“At this point, she stops an angel of death from murdering the kindly, but irreligious, studio head”). Basically, a good answer makes your novel easier to write.

But that answer could just as easily never come. Or you could get so tired and frustrated that you convince yourself that a bad solution is actually a good one.

Because at some point, the good fear–the fear that lets you know you’re making a mistake–turns into the bad fear. The bad fear actually gets stronger as the novel gets better. For me, the bad fear is mostly a voice that says, “You’re gonna fuck this up. You’re gonna lose this.”

And it can be overpowering.

The bad fear is the reason why, when I’m on the threshold of really getting into a project, I’ll sometimes spend days (or even weeks) not working on it. Because if you don’t work on something, you can’t fuck it up.

The truth is that the bad fear isn’t wrong. Sometimes I do screw it up. And sometimes I’m feeling really confident about a project but then, the moment I start to work on it, I realize that it everything has evaporated. But, unlike the good fear, the bad fear is unproductive.

The good fear stops me from making mistakes; the bad fear stops me from doing anything. The way I’ve written this blog post, it sounds like the two fears are very separate. But that’s not really the case. They feel very similar. And oftentimes they coexist. The differences between them are very subtle. Generally speaking, the bad fear tends to abate the moment I begin writing. Conversely, the good fear builds and builds as I write, until it eventually ejects me from the story.

That’s why I can’t really outline stories. The bad fear loves outlines, because they feel like work but, on the other hand, you can’t really screw up an outline. And the good fear doesn’t even seem to operate on outlines. I can cheerfully draw up the most incoherent outline in the world without getting even a twinge of the good fear. Basically none of my instincts really come into effect until I start writing actual words.

Anyway, this has been your monthly dose of semi-mystical writing advice. I wouldn’t advise putting too much stock in it. After all, I’ve written a bunch of novels, but none of them have been published. It’s entirely possible that the difference between my “good” novels and my “bad” novels is something that’s apparent only to me.