Tuesday, 26 April 2022

How Your Favourite Authors Cheat, Why That's OK, and Why Every DM Should Too

[Warning: this entry contains mild spoilers for A Song of Ice and Fire and The Book of the New Sun.]

When reading an intricately-plotted novel, it is easy to fall into the trap of feeling intimidated by the author's great intellect and technical skill. How can this author have planned this all out in advance?

The truth of the matter is, it is almost certainly the case that the story was largely made up as the writer went along. (If memory serves, Bernard Cornwell and Kurt Vonnegut are notable exceptions.) They have followed Humpty Dumpty's advice and begun at the beginning and kept going until they have got to the end, and then stopped. The intricate plotting really just comes from the author using what he has already written as seeds for future plot developments ("I know, wouldn't it be interesting if the lady in the red hat from way back in Chapter Two turned out to be the sister of the dwarf who I've just written into Chapter Nine?"), and then going back and ret-conning things later to keep things tidy. It's a product of being constantly attentive to what one has written and how it might tie into what one is writing, or imagining, now - coming up with interesting premises, situations and characters and then making connections between them as one goes.

Think, for example, of Dr Talos and Baldanders in The Book of the New Sun. I don't care how much has been read into that book, or Gene Wolfe's writing process, but you will have a hell of a job convincing me that those two didn't just appear as interesting figments of the imagination to begin with ("Wouldn't it be cool if Severian ended up in cahoots with a fox-like playwright and his giant companion?"), who Wolfe later on decided to bestow with much greater significance during the course of the writing of what became The Sword of the Lictor. He didn't plot out the entirety of their role before sitting down to begin the first sentence of The Shadow of the Torturer. He wove it into the telling of the story as it unfolded in the course of the writing. Once it was clear who Baldanders really was, and he had completed his first run-through of what would become the final version of the book, Wolfe then presumably went back and made all the necessary amendments in the rewrite so you couldn't see the "join". 

Think also of Tyrion killing Tywin Lannister towards the end of A Storm of Swords. Did George RR Martin have it in mind that would happen at the start of writing A Game of Thrones? Almost certainly not. The idea that Tyrion might end up killing Tywin may have crept up on him gradually over the course of writing the first drafts of the books, or occurred to him in a flash of insight, but it would have been something that emerged from the story - and how the relationship between those two characters, and the characters themselves, had developed in Martin's mind - during the initial writing process.

What seems like very clever plotting, in other words, is clever, but it is really better described as clever rereading and rewriting. As readers, it is very easy to forget we are not reading the story as the writer wrote it. We are reading the final product of a long process of rewriting and editing: the final version that is presented to the world, not the first draft that was 100,000 words too long and will forever remain locked in the author's attic.

It helps to bear this in mind as a DM. The idea that one could plot out "an adventure" in advance (except for a very simple and boring railroad), or could fill in all the details of a campaign setting before play commences, is a pie-in-the-sky. As with writing a novel, DMing is really an iterative process - it's just that while an author merely riffs on his own ideas, a DM can also riff on those of the players. Events transpire not because they were carefully detailed back before the start of the campaign itself. They transpire because the interesting locations and NPCs the DM has come up with, and the PCs' interactions with them, bring about connections in his mind. It's not that Steffi the Orc was intended to be the PCs' arch enemy all along. It's that Steffi the Orc was captured by the PCs after the rest of her companions were killed in a random encounter, and she was then released, and the DM thought, "Wouldn't it be interesting if she decided to get her revenge? And wouldn't it be nice if she teamed up with Cedric, the hireling the PCs kicked out of their party for stealing, after finding him wandering in the wilderness? And wouldn't it be good if the two of them decided to burgle the PCs' treasure stash while they were away...?"

The exercise comes, in other words, not from pre-plotting, but plotting as a continuous process or flow.

[I am running a Kickstarter throughout April. You can read more about it, and back it, here.]

[I also did a nice interview with the ever-interesting Solomon VK of Worldbuilding & Woolgathering here. You will find it relevant to your interests.]


  1. The other angle is when stuff happens in later books people go look through earlier books for foreshadowing (for example there is some STRONG foreshadowing of the Red Wedding in the preceding book) but people don't look for foreshadowing of things that never happened.

    For example in the first book there is some foreshadowing of King Jaime which obviously went nowhere and then there's the example of Taim in WoT whose future was foreshadowed heavily enough that readers guessed it, which made the author get pissy and change it.

    When looking at foreshadowing it's also hard to tell what stuff played out as intended and which stuff was repurposed and came to be linked to different things than was originally intended.

    1. Yes, I guess my point is that what looks like clever foreshadowing may very well be something that was added in the rewrite once GRRM came up with the idea of the Red Wedding in the course of writing. But, as you point out, we can't really know.

    2. Eh, I think for this one it's just too explicit:

      From ACoK (the book before the RW):

      Patchface: Fool's blood. King's blood, blood on the maiden's thigh, but chains for the guests and chains for the bridegroom, aye, aye, aye.

      The Wood Witch: I dreamt a wolf howling in the rain, but no one heard his grief. I dreamt such a clangor I thought my head might burst, drums and horns and pipes and screams, but the saddest sound was the little bells.

      Danny in the House of the Undying: She came upon a feast of corpses. Savagely slaughtered, the feasters lay strewn across overturned chairs and hacked trestle tables, asprawl in pools of congealing blood. Some had lost limbs, even heads. Severed hands clutched bloody cups, wooden spoons, roast fowl, heels of bread. In a throne above them sat a dead man with the head of a wolf. He wore an iron crown and held a leg of lamb in one hand as a king might hold a scepter.

      Theon's dream: And then the tall doors opened with a crash, and a freezing gale blew down the hall, and Robb came walking out of the night. Grey Wind stalked beside, eyes burning, and man and wolf alike bled from half a hundred savage wounds.

      That's just too many things pointing too specifically in the same direction. There's off course buckets more in ASoS but that doesn't count in the same way since it's in the same book as the RW.

  2. I'm reminded also of those authors who have a number of scenes or moments they want to portray, then link these together somehow. CS Lewis's Lamppost in the snowy woods is the example that sticks in my mind. I understand that Philip Pullman will sometimes plan stories using sticky-backed notes, so as to rearrange elements easily.

    'Ever-interesting'? Maybe. But the interview was certainly a lot of fun and there are still questions unanswered!

  3. Yeah, i couldn't finnish ASOIAF exactly because this mysterybox writing doesn't read like a great plotting to me at all :D
    I mean, it can be an useful tool, especially for gamemastering, but Martin is, to me, more of a cautionary tale.

    1. He definitely took it too far, to the point where the connections between all the different seeds he’d sown became unmanageable….

  4. Very good points.

    One way of doing this procedurally is to constantly write fresh rumour tables (in the style of Keep in the Borderlands and its ilk). I find that this is a great way to keep the campaign 'alive' - and to spin off plots from events at in previous sessions. So, rumours might concern an NPC recently encountered or a location recently explored - or provide a link between the two, among many other things.

    And because rumours aren't necessarily true, the GM has lots of room to allow plot threads to develop or die. Something the PCs ignore might just wither - or it might grow legs if some other development appears to cast a new light on it - whether in the mind of the players or the GM. And a rumour that preoccupies them is bound to give the GM all sorts of ideas.

    An aside: something my players enjoy is when they recognise a rumour as a distorted account of their own exploits - especially when they don't work it out immediately but realise at a later point.

  5. I saw a table Rowling created for one of her books showing what happens to each character in each chapter. Of course it's impossible to know if that was a pre-draft table or one created near the final polish.

  6. First, congrats on the Kickstarter, which I of coursed backed and I'm quite excited for. If you do more I'd love to contribute.
    So, I agree and disagree with what you're saying.
    As a writer I can't imagine not outlining/planning things out. In fact, I'd say that my transition from aspiring to professional writer coincided with when I started seriously outlining (even though I don't find it to be tons of fun, my least favorite part of writing other than rewriting).

    Some caveats:

    First, while I tend to write a detailed outline, that doesn't necessarily mean I stick to it. In fact, my favorite part of writing is when characters start "telling me" they want to do things that I didn't plan, and I take the narrative in that direction. But having an outline is at the very least a safety net to make sure I don't write myself into a corner or lose steam midway through.

    I've been teaching writing for over a decade, and while I try to give my students as much freedom as they want, it's rare that something that's not well planned is going to wind up any good.

    Living in NY and LA most of my life, I've met many people that call themselves writers. I tend to not be very judgmental as to what makes someone a writer. Some people put the benchmark at getting paid. Which is certainly something to be proud of, but I think the bigger hurdle (besides actually sitting down and writing rather than just talking about writing) is FINISHING something. And in my personal experience not planning to some degree is what prevents aspiring writers from finishing. They get stuck in the second act and have no idea where to go.

    To the extent writers succeed without planning, I think that's due to the innate story sense we all have. At this point we've all read, watched and otherwise consumed so many stories that we can't help but internalize some kind of structure. I don't think that's enough for most writers to succeed professionally, but it's more than enough to run an RPG campaign.

  7. (Evidently my comment was too long, so the 2nd and 3rd part follow in separate comments - apologies):
    Another caveat to my strong feelings about outlining/planning. is that I'm not a novelist. I'm primarily known for writing comic books and graphic novels, and I've also written for film and TV. With film and TV there's a very specific structure that you need to adhere to, and if you don't not only will you be marked as an amateur, but no one will give you money. With TV and film there's so much money involved that the powers that be aren't generally going to just let you do your thing unless you can show them where it's going.

    There are exceptions to this. LOST is a good example - I've heard different things from people who worked on the show, but I'm convinced they had no plan for how it ended until at least midway through the series. That was from a different era in television though, and I don't think they ever foresaw it going as long as it did.
    I'd argue the Star Wars sequels are another example of this. There was never an overall plan for the third trilogy, it was just one director passing the baton to another. I'd argue that's why (in my mind) they were not very good.

    Writing prose gives you more freedom. Structurally, most short stories don't need more than a beginning, middle and end.When I've written short stories I've been able to "just write" without much
    planning. It's one of the things I enjoy most.
    I've never written a novel, so I don't know if I could keep that going. I believe most of my novelist friends "genre" writers (crime, sci-fi, horror etc.) who are successful on an entirely different level than me outline to some degree. I apologize for being too lazy to ask them now.

    I don't know whether GRRM outlines. I do know he had to turn in an outline of the books he hadn't finished to HBO, which is what the showrunners of Game of Thrones followed.
    He was more involved in the series early on and expected to finish the novels before the series concluded. So that outline may have been preliminary.
    I don't think he was thrilled with how the show ended, or at least how the ending was received. It's possible that had he gotten that far in the books, he may have made major changes (or simply executed on the outline better). It's also possible he would've done much the same thing. It's easy to see the flaws in the last season or two now, when the whole world has weighed in. I think how the show ended/was received has made it even harder for him to finish the series.

  8. All that said...I think what makes gaming so unique is the improvisational nature of it. You'll get no argument from me that the best adventures and campaigns are sandboxes, or at least have a sandbox element to it.

    I'm primarily a GM, and my favorite part of that role is when my players do something I didn't expect, and I try my best to take the campaign in that direction. It's I suppose the gaming equivalent of having characters "tell you" they want to do something you didn't plot out, and for me it happens more frequently than in my writing.

    This is part of why I love Yoon Suin so much. Among many other wonderful qualities, the procedural nature of generating my unique version of Yoon-Soon makes it a surprise even in the planning phase. I think the basid of creativity is trying to make two things that seemingly don't fit together, whether's it's a seraglio and cockroach men in Yoon Suin, or mixing genres or characters in a movie or novel.

    Do it well enough and a certain type of story writes itself. A campaign doesn't need to follow three act structure (and neither to picaresque stories). I'd love if there were some sort of tool that did for writing screenplays, comics etc what it does for gaming.

    Alas, there's precious few tools on the level of Yoon Soon in RPGs.
    The group I run are in love with 5E, and they are such wonderful players and friends that I run that system. 5E is - at least for me - as system that's harder to improv for that B/X based systems. Especially as my players have gotten use to the bells and whistles of what the official WotC modules offer on Roll 20 in terms of system automation, beautiful maps with lighting built int etc. Combine that with the fact I feel like I only have so much creativity that I can devote to gaming vs. the writing that pays the bills, and I'm ashamed to say I rely on their published adventures more than I'd like.
    (In WotC's defense, I'd say Rime of the Frostmaiden does a pretty good job of being a sandbox in it's first two acts, before succumbing to some degree of railroading at the end).

    Last, I'll mention that I'm in the midst of playesting the 2nd adventure I've written (likely to be the first one published). It's hard! I say that not to discourage anyone from trying, more to say I respect good game writing even more. But one of the things I'm trying to do is find just the right balance of structure and open play. I'm doing my best to lean hard into the latter, but there's an art to giving players enough to run with while making sure there's the opportunity for a satisfying ending. (Let alone balancing giving the GM enough info to run the adventure without overwhelming them etc.)

    Sorry for such an overly long response. TLDR - I absolutely agree that you don't need to intricately plan an RPG campaign, and that doing so - and sticking to that plan - can be detrimental.
    I'm a little less sold on the idea one can write a novel without some degree of planning. If an aspiring writer were asking for my advice, I'd tell them to outline even if they don't stick to it. Of course, the best way to get better at writing is to write, so if the idea of outlining is an obstacle, better they write without an outline than not write at all.

    1. Nice to hear from you, Mark! Yes, I was going to say exactly what you said - I think novel writing is different from comics and film/TV writing because it is so (relatively) free and also usually much longer, as well. It seems to me that almost all novels are written from the starting point of the novelist having an initial starting idea and then writing the story to tell it to *himself*, as a first draft, and then finessing that draft until it's in publishable form. I am sure that there is an interative process of outlining, and re-outlining, for most people as well, though, so I don't think planning is entirely absent.

      I say this as somebody who has written, not novels, but a proper published academic book, and lots of long-form academic articles. For me, the process is one of having an initial idea, a feeling about where it is going, and a very rough outline, and then constantly shaping and re-shaping the outline as I write (indeed, until the very end). I can't imagine knowing what I am going to write before I write it, because discovering exactly what one wishes to say happens through the act of writing itself. I think of it as being a bit like sculpture. You can only do it by picking up the hammer and chisel, and you can only really discover what you want to produce by doing the chiselling. The two things - planning and producing - happen simultaneously, if that makes sense.

    2. Great to hear from you as well! It's funny, I just listened to a podcast with Don Winslow, my favorite crime novelist, who wrote the Cartel trilogy etc. and he said he doesn't outline.
      Even when I write screenplays I reshape the outline - for me it's a kind of security blanket to have it even if I know it will change.
      The kind of sculpting you describe reminds me of a book on writing I quite enjoyed this past year. It's called "A Swim in the Pond in the Rain" by George Saunders. He's (evidently ) a well known short story writer, and he also teaches writing and Russian literature. The book is a distillation of his course. It alternates between classic Russian short stories by Chekhov, Turgenev, Dostoevsky and Gogol, and then essays by him explaining how they work from both a writer and reader's perspective. I find most books on writing to be rubbish, but I'd recommend this one (if only for the Russian short stories...they are classics for a reason).

    3. I've heard of that book and I think heard an interview with Saunders. Will have to track it down!

    4. When I write comic/graphic novel scripts, I also work from an outline (usually after writing up a precis of the story), but that's primarily because pagination is so central to the (visual) finished product, so it really matters that certain content appear (for example) on the same page or spread. Though I don't have any experience with it, I would imagine that writing for the screen imposes similar strictures.

      With anything long-form (novels, etc.), however, I don't outline at all, nor do most--but not all--of my writer friends and acquaintances. The early drafts are pure exploration, and are typically heavily changed/rewritten a number of times. (During which there's plenty of time for fashioning the joints, so to speak.)

      With academic work, of course, I (like most) work from a sketchy, dynamic outline. To an extent, that's because of the nature of argumentation: one doesn't typically start a writing project with only the pieces understood, and from there construct the major claims; rather, writing starts at the point where the major claims begin to take at least inchoate shape, and then are refined through the drafting process.

    5. Yes - "making to know" - see below!

  9. Wodehouse:

    “Writing my stories — or at any rate, rewriting them—I enjoy,” he said not long ago, adding:

    “It is the thinking them out that makes the iron enter into the soul. You can't think out plots like mine without getting a suspicion from time to time that something has gone seriously wrong with the brain's two hemispheres and that broad band of transversely running fibers know as the corpus callosum.

    “I always have to make about 400 pages of notes before I can get my scenario set, and there is always a moment, as I read them over, when I pause and say to myself, Oh, what a noble mind is here o'erthrown.’ ”


    1. Hmm. Interesting, though I wonder if I really believe that. 400 pages of notes?

    2. For an approximately 200 page novel, yes, also the notes were handwritten, they looked like this:



      In Dickens rough drafts you find him similarly asking himself lots of questions.

    3. That "asking oneself questions as one writes" is probably extremely common. That's how I do it.

  10. There are many different writing processes. One screenwriter and script doctor named Jeff Kitchen outlines backwards. He starts with very rough concepts, and decides on the major themes, dilemma and ending and then works backwards to the beginning of the screenplay.

    Novelist Steven Pressfield uses a process he calls the foolscap method. It's a simple technique where he works out a sin
    gle sentence for each major sequence in the book;the instigating incident, the middle build and the climax.

    Other writers get an idea and start typing.
    Many different processes.

    1. That's certainly true. But I would reckon "writing the first draft to discover the story" is most common.

    2. I think maybe we are saying the same thing just in different ways. A big part of the work is discovering the story. Some writers, perhaps most (I don't know), start writing the story and see where it goes and then revise and rewrite. Others write out story beats on index cards and play with them and then once they have a sequence figured out go fill in the details. All that to say they have to "Make to know." a phrase I picked up from a book of the same name I reviewed a few weeks ago.

      The thesis of the book is that creators (no matter what art or design problem they are working on) have to find an entry point and then just make something to figure out what the final thing looks like. The product doesn't just appear in their head and then get made. They have to make it and discover it through some process.

    3. Yes, you're right. I really like the idea of "making to know" - that captures it perfectly.