Questionable: How Do You Start a Book?

K asked:
And when you sit down to start a new story, do you plot out major scenes? (know them ahead of time) Or do you start with an idea and the ‘aha’ moment at the end? In other words, do you know where you’re going when you start? Or do you just start with two characters?

There are a million ways to start a book, and all of them are the right way if they work. My way is the least efficient, so please note, I am NOT recommending the following as a path to follow.  It’s a grossly inefficient way of writing a book. It’s just only way I’ve got.

People start talking in my head, sometimes inspired by an idea, a movie, a snatch of conversation, a news story . . . damn near anything, really.  Usually, they just wander off and I never think of them again.  But sometimes they get interesting and I start writing down some of the dialogue because I love writing dialogue.  Actually, I don’t write dialogue, I just write down what the voices in my head are saying; there’s a conversation in my head and I’m doing the transcription.  

Then when I get to twenty or thirty thousand words of inane chatting, I start to think there’s a book there and I get serious.  Who’s the protagonist?  (I usually know that one, didn’t on Faking It.). Who’s the antagonist?  (I never know that one.) Who’s the love interest? (I usually know that one, but not always.) What’s the conflict? (What conflict?  These are just verbal people.)  And answering those questions makes for more conversations; those of you who have seen my first drafts know they’re basically radio plays. 

That wandering around and writing down random conversations I hear in my head goes on until I have enough scenes that I go back and put them in chronological order and then keep writing until I have most of the discovery draft done.  

Then I step back and go into analytical mode—where’s the conflict, where are the act breaks/turning points, how are those character turning points, too, how does the character arc, and so on, ending with “What is this book about?” 

 And then I rewrite until I get a complete draft with turning points done, and I take a step back and look at it and see that I wasn’t writing the book I thought I was writing at all, it’s a different book, and the characters have things going on that I missed and need to be deepened, and I go back in and rewrite some more.

Lather, rinse, repeat.

BUT that’s how I have to do it, it’s not the way I’d recommend anybody else do it. Some people can plan their character arcs completely ahead of time which is much more efficient.  I have to find my way through the dark first.  I think that’s fairly common; Jo Beverley used to say that writing a book was like driving through a fog, you could only see as far as your headlights, and I think that’s true for a lot of us.  But there are also a lot of writers like Bob Mayer who map the entire book out first and then write to the map.  Much more efficient. I just found out that Rex Stout wrote one draft of all his stories, started at the beginning and went on until he got to the end. I think his pre-planning was massive, but once he started writing, he just wrote the book, no rewriting.

In the end, it doesn’t matter. What does matter is having a story that you have to tell, that will nag at you if you don’t tell it. It’s the story you must tell that matters, not where or how you start it.

To show you how inefficient my method it, here’s four years of first scenes of Nita.

Four years, people.

Really, don’t do it like I do it.

42 thoughts on “Questionable: How Do You Start a Book?

  1. I was really enjoying studying the evolution of the scenes (although I have a huge stack of papers I really ought to be grading, but they’re for a class where studying an author’s writing process could be justified, which is how I managed to justify indulging myself like this), but when I got to the most current one, it asked for a login, and my normal WordPress login didn’t seem to work. Is there somewhere I’m supposed to create an account? I’m sorry if I’m overlooking something obvious.

    1. It shouldn’t ask for a log-in.
      I wonder if I put the wrong link in. The only person it should ask to log in is me.

    2. I love this more than I can say. I feel like scene revisions are a really weird thing to get all goofy over, but this truly makes my day, or, more likely, my week. (In fact, although this is my favorite so far, I’ve been enjoying all of the Questionables–thank you for launching this!)

  2. A couple of years ago, I interviewed Alexander McCall Smith (No.1. Ladies Detective Agency and 44 Scotland Street series) who is wildly prolific (2000 words a day come rain or shine) and he spends a good deal of travelling/walking time but when it comes to it, churns out the draft and then does very little editing. He started out life as a law professor, so that may explain both his prolixity and his strong internal editing mechanisms.

    Personally speaking, I have a whole series of potential opening incidents and an end in sight, but not always sure of the pathway between the two. The thing I do try to get right is timeline, broken down into years, months and occasionally weeks. That gives me basic structure.

    1. Ann Bridge said in her memoir that she, as the wife of a diplomat and mother of a family, didn’t HAVE a lot of time to write, so she (approximate) did rough drafts in her head and when she was at her desk, had most of the preliminary stuff mentally organized.

  3. I clicked on the link to read the different versions of the scene. I am hopeful that we will be able to see the 20119 version of the first scene of Nita but when I click the link, I am prompted for a username and login. Just in case you intended us to read the latest version, I wanted to let you know that it wasn’t connecting.

    By the way, I am watching an episode about the Bronx Zoo where it focuses on a flamingo that has blown its ACL. After surgery the zoo keepers put a mirror so the flamingo sees another flamingo and play sounds of a flamingo flock to keep the recovering flamingo calm. The keepers discuss how flamingos are flock animals and need the presence of other flamingos. Shades of “Agnes and the Hitman”! Your research is impressive.


    1. Okay, I’d linked to the dashboard page, the one I edit, instead of the published page. SORRY. It should be fixed now.

  4. Jenny wrote, “There are a million ways to start a book, and all of them are the right way if they work.”

    In “In the Neolithic Age,” Rudyard Kipling wrote, “There are nine and sixty ways of constructing tribal lays, / And every single one of them is right.”

    Great Minds Think Alike

  5. It took me a long time (and a lot of false starts) to realize that I love starting a new story so much that if left to my natural instincts, I start writing before I’m ready, and then I get stuck. Now, I have a list of seven things (my “rule of seven” instead of “rule of three”) that I need to get from the voices in my head before I start writing, and I usually do some plotting first for turning points and whodunnit (since I write mysteries, and not all mystery writers need to know the culprit before writing, but I learned that I do need to know). Otherwise, I’ll write the first chapter in a rush, maybe the first three chapters, and then I realize I don’t have enough stuff for the story, or I’ve dumped everything that needs to be fed out in the book, all in that one to three chapters.

    If anyone’s wondering, the seven things I need to know before I start to write: a title (although it often changes, and that’s okay), premise (in a very few words, often including the setting), a subplot (often a romantic element), a non-mystery-solving goal for the protagonist (like harvesting the garlic in the garlic series), a theme or motif that carries through the series (like health challenges in the Helen Binney series), an arc for the protagonist, and something interesting about the killer. When I start writing too fast, all I know is the premise and the protagonist and setting, and for me, that’s not enough.

      1. Not available until next spring. Just got initial edits. I don’t think it’s even got a place-holder anywhere. First book is SIX CLOVES UNDER (A garlic farm mystery), for release in March (or maybe April) of 2020. Second book, RHUBARB PIE BEFORE YOU DIE and third book, LAID OUT IN LAVENDER, should follow at six-month intervals.

        The series is set on a garlic farm (and a rhubarb field gets added in the second book, and there’s already a lavender field for the third book), featuring an apps developer who inherits her aunt’s farm, only to realize the supposed accident was actually murder and no one believes her, so she has to figure it out herself while also saving the garlic harvest, despite having zero agricultural experience. A major supporting character is Pixie the cat.

        And there are recipes! Although, I have to admit I was stymied initially when my agent asked me to add them to the manuscript, because mostly I just chop up some garlic before I decide what I’m having for dinner, so it goes in everything and there are no actual recipes.

      2. I don’t want to nest too deep, so this is a response to Deborah (and everyone who hearted her comments) — thanks for your interest!

  6. I don’t write anymore, and never wrote all that much. Several short stories are about “the chocolate couple.” I don’t think I ever named the guy. The first sentence of every one of them is “Jeanine said one of us needs to go to the store”. Every one. The punctuation changed, sometimes but the words didn’t. And the person who always seemed to be the “one of us” was the narrator, Jeanine’s husband, and the reason was almost always chocolate.

  7. Oh Lord and Lady, God and Goddess.

    Having read the others before, I went straight to 2019. I got to Witherspoon getting into the car and I just started grieving Mort. I’m only half joking.

    Will read the rest when I’m ready for it. Argh.

    Thank you for posting, though.

  8. I usually (but not always) outline before writing, although sometimes the first couple of chapters get done first, if an idea is very loud. Like Gin, I have certain things I try to have set down before I start serious writing. I usually have the theme (although that sometimes shows up at the end of the outlining process when I can look at the whole story). Something like: connections with others give purpose to your life.

    I know the main characters: both protagonists (usually two) and antagonists (one or two–in the romances, there is usually one for the female protag and one for the male), and their basic motivations. General conflicts between the protagonists and the antagonists, along the lines of “what does X want, and what does Y do that stands in the way of X getting it?”

    In my urban fantasy and paranormal romances, I have very detailed outlines which cover most of the action, although things do crop up or change in the course of the writing. My characters have been known to do the unexpected. My contemporary romance and Women’s Fiction, on the other hand, I may just have a 3-5 page summary–they don’t seem to lend themselves to outlining as well.

    If I can, I even write up what would be the back cover blurb, because that gives me a good idea of what the book is about. Mind you, all this is best case scenario. It doesn’t always work that way.

    1. Oh, that made me think of something. I don’t usually do fantasy that requires worldbuilding, but I actually have one UF manuscript that I play with occasionally, so for anyone doing fantasy, having a plan for developing that world (what the magic is, what it does, what it doesn’t do, who can do it, that sort of thing) is useful, and you’d want to figure out whether you should work it out in advance or as you go along. I’m not sure how Deborah does it, but she might have some suggestions.

      Martha Wells, who does amazing worldbuilding as everyone here has found out in the Murderbot series, but even more so in some of her fantasy books, has some tips at her website on worldbuilding. At least, I think that’s where I found them when I was working on my UF.

      1. Absolutely. In my Veiled Magic series, I purposely used a number of unusual paranormal creatures (all hidden from human society except Witches, who had come out of the broom closet). That meant things like figuring out what they were, what they could do, which ones could product offspring with each other, and what traits those offspring would have (ditto halfbreeds with humans).

        Because this was at the height of the vamp and shifter days, I purposely had none of the first, and the “shifters” were Ulfhednar, berserker shape-changing warriors from Norse mythology. Then there were witches, Fae (like the unbelievably beautiful, but cold and self-centered actors you sometimes see who never seem to age and who live on the energy of adoration), Ghouls (that gray man always sitting at the end of the bar soaking up the bad vibes of everyone around), and dragons. Because dragons.

        I did MASSIVE amounts of world building before I started writing. The series was based on a short story, in which I had established a bit of the set-up, and it was set in a world like ours but a few years in the future, so that at least made things easier.

  9. I miss Mort but it’s tighter now and the mystery texter adds something to Nita’s reason for being there. I really like the “No one should be afraid on my island” bit. It tells us a lot about Nita.

    1. Yes, I love that the mystery texter adds that sense of urgency, too, for why Nita would head out in her pyjamas while still drunk. If she thinks someone is in danger, that makes sense.

  10. I’ve mentioned it on Arghink before, but my big epiphany was that I write like I crazy quilt – bits all over the places, scraps of scenes, scraps of dialogue, and I shouldn’t try and shape it into a cohesive narrative until I have far more fabric than I think I need. Forty sodding years it’s taken me to work out that I can’t and shouldn’t try to write a novel-length work in a linear way, because that kills my writing deader than dead. Forty years. I’m a slow learner.

    1. I’m similar in that I also end up with a corpse of a story if I write the first draft linearly. However, I do tend to outline pretty heavily, even if things change as I go. I jump around while drafting, writing to relieve the itch of the moment, until I have the bulk of the story on the page. Then I go back and write any necessary transition scenes before starting to revise.

      It may not be efficient, but at least the words have life in them.

  11. I’ve had too many stories end around page 70 to not do at least some outlining. But I often only outline the first 2/3 or 3/4. Since there’s always a happy ending in my books, I know how it is going to end – just not how it is going to get there. So if I give myself a running start with the partial outline, I’m still leaving myself some room to let the characters make their own decisions and take control of their plot. I can be too rigid sometimes, and need a reminder to let my stories breathe on their own.

    I have to have the first scene pretty set in my mind before I sit down to start writing. That’s not saying it won’t change – but I need to have that scene before I can even get going.

  12. I had to respond to this one.

    By way of introduction, I am the husband of a longtime lurker, and am a Jenny fan in my own right; the Dempsey Five Step Swindle is Holy Writ in our household.

    About a year ago, after more than 40 years of failing to finish a novel, I did an inventory of my writer’s tools, and made a list of things I was unusually good at, and things I really enjoyed doing, and then set out to design a story around the things that were near the top of both lists; “dialog” was number one on both. Six months later, I had finished a novel; five months after that, it was up on Amazon (and elsewhere). Not surprisingly, it is (at 99%) one of the most dialog heavy books in the history of English. The second draft consisted of changing fewer than five hundred words in the name of “smooth”, and contracting about 2000 eligible word pairs.

    As Jenny says, every process is different, and the one I used this time isn’t repeatable. But sometimes it’s a good thing to be drowning in dialog.

  13. Completely off topic, The Goblin Emperor is on sale for $1.99 on Amazon right now, if anybody was thinking of trying it after all the recs here (I still haven’t read it, so not recommending, just saying $1.99).

  14. Mort. (sniff)

    Maybe when I stop missing him, I’ll see how that first scene – the book – improves without his presence. Although already I discern it kinda does.

    Mort. (sniff)

    1. I know, I loved him, too. He just wasn’t pulling his narrative weight.
      It strengthens the scene because I need Nita isolated there, and Mort was such a constant back-up for her. He was a great brother, but a lousy narrative decision.

      1. Nita definitely feels more alone. I love the way Frank and Jason both respond positively to Button and not to Nita, that really underscores Nita being alone. And the mystery texter works for me too. I like her protectiveness.

  15. At work so don’t have time to use the link yet, but had to say – sure doesn’t feel like 4 years. Of course, I didn’t do any of the work, either.

    1. You did some of it, reading those drafts and all my rants. The Argh People had to put up with a lot of whining.

  16. I’m just working on my first historical which is set around 1860 in Germany, so I need a lot of research for that. My first impulse was that I have to do all the research reading before I can get started with my story but it didn’t lead me anywhere at all. There is so much that I might need to know!

    So I tried the other way around and started to write the scenes that I already knew I needed. I wrote 33 pages in a couple of days. I might have to change some stuff once I’m sure about the historical background but at least I’m getting somewhere – and I’m having fun, too.

    Are there any writers of historicals out there? How do you do it?

    1. Ron Carlson, one of my best cw teachers, said he was writing a scene set in the British Libray and he stopped and thought, “Wait, I don’t know anything about the British Library,” and then he told himself, “Don’t look down,” and kept writing because he could always look up the British Library later. That’s my mantra while writing the discovery draft: Don’t look down.

    2. I think you’ve chosen the right path: it’s so easy to spend years researching, and lose all the juice of the story.

      1. On the other hand, if you can get the balance of first draft to research right, the research can hand you a whole lot of story. I’ve just spent ages piecing together info on a particular place, but the moment I had that, I could walk the space in my head and a whole lot of stuff I was trying to write suddenly made sense and I could see the action. And then I found something in the photos filled in a gap in the plot

        1. Research on other things also gave me, by pure accident, entire characters and the path into their backstories. But it’s very easy to get too distracted.

    3. I’ve written two historicals and have a WIP that is not even back-burnered, it is, like, in the freezer waiting for the contemporary stuff to stop coming so fast.

      My first started life as a project for a master’s program seminar. I don’t even remember what the assignment was, but what I turned in was an lengthy essay on the microeconomics and household of a Bordeaux winemaker in the years just before the French Revolution.

      Then I had all that world-building sitting around, and ended up writing a novel about the daughter of that winemaker. Once the first one was done I had the hero’s best friend who could operate in my fictional world, and I knew who he was and what his issues were, and I wanted to sort him out, so he got a book.

      That second book precipitated the currently-hibernating book 3. But more pertinently, the first one started with a thick stack of research that happened to engender a character, who came to life due to the necessity of giving that research a narrative.

      Book 3 also required a stack of research because it is set not in France or England, but in Ceylon (now Sri Lanka). WWIT. I wrote some scenes for that before even finishing book 2 because it was clear to me that a certain character was going to carry his own book. He is a piece of work.

      My method for organizing the historical research (so as to not go down the infinity of rabbit holes, history is FASCINATING) is to start with the character. What does this character do, how is that done, where does s/he live, how does s/he dress, what does s/he eat, now find out what’s factually accurate that will answer those questions and stay away from the edge of the cliff. 🙂 Generally speaking in historical romance I think the less obvious the research, the better* for the story. People are people; a little scene-setting goes a long way. For me, the story also starts with the character. Without a good character, who cares about the story?

      *exception: modern speech in a historical setting. please god no.

  17. I’m not a fiction writer, but I’m convinced that some of my favorite authors start with a problem and work it out. Like Connie Brockfield in My Dearest Enemy, I’m pretty sure she started with how do you write a romance when marriage was so unjust for women but having kids out of marriage was so unjust for the kids…and then wrote the book to work itout.
    When I write nonfiction, I start by knowing what point I want to make and go from the start to the ending (and then edit), so maybe that’s what I assume everyone does.


Comments are closed.