This is complicated by the fact that one of my best friends and story-noodlers is highly character-centered and dislikes having to make up much/any deep background in advance of the story. So her noodling questions have all been focused on the characters and plot (because she knows I do plot), which sometimes hits the “deep worldbuilding” button, but mostly doesn’t.

What I mean by “deep worldbuilding” is all the background, from geography to cultural history, that shapes the place and time the characters are living in. When I’m writing alternate history, I have many libraries’ worth of information to use or choose not to use. I can look up where the rivers and active volcanoes are, or where certain crops originated; I have the Han Dynasty, the Egyptian Pharaohs, and the Greco-Roman Empire that can be assumed with all their real-life consequences, or tinkered with, or eliminated with all those consequences (what if the rulers of the Indian subcontinent had chased Alexander back home and conquered Greece? Or Cleopatra had managed to annex Rome, instead of Rome getting Egypt?). Even if I made the world unrecognizable save for the geography (because, oh, aliens messed around with life on Earth at the end of the Mesozoic Era, so we have civilized dinosaurs instead of us), I’d know where the mountains and rivers were, and what the climate was like in various places, and so on.

But the New Thing isn’t alternate history in any way, shape, or form. Bits of it are modeled on Real Life History, but it’s more like visiting a museum exhibit of Michelangelo’s work and then coming home and trying to build a Cubist version of the Pieta out of cardboard boxes than it is like a mash-up of actual places and events.

Doing a lot of deep worldbuilding in advance is not for every writer, but it helps me. In fact, as has become quite clearly obvious, I need to know a fair amount of it, or I can’t get things to hang together properly at the immediate-backstory stage. That doesn’t mean I do all the deep worldbuilding at once; on the contrary, it develops in fits and starts, forwards and backwards. That is, sometimes I know something (like “this is a coastal city”) that implies a bunch of other things (a harbor, trade, seafood dishes). Other times, I know something (there are three distinct and mutually exclusive types of magic) and it begs a question (how were they discovered, and why do they have more-or-less equal status and emphasis?). The answer to that (three major empires back in their early history, each with a different attitude/philosophy toward What Man Is Allowed To Tamper With) implies some more things (my city must be somewhere that was either not directly influenced by any of the empires, or influenced equally by all of them, there are going to be at least some people who still have very strong opinions about whether each type of magic is good/bad).

I like the idea of a trade crossroads at some point in the middle of my three empires, which fits with the harbor-and-trade part I established earlier, but it might be inconvenient. I’ve already got a three-way magical conflict; do I really want a three-way philosophical and political conflict as well? Even if it ends up being just the historical remnants of the empires that my present-day people have to deal with? On the other hand, can I really avoid it, given what I have so far for background, even if I stick the city far away?

If I make the location somewhere well away from the ancient empires, then it’ll need to have some local resource that’s valuable enough to stimulate trade with all three, but not so valuable that any of the empires would come all the way out there and conquer the place to get it. So not iron or gold, but maybe silk or purple dye or porcelain. That will probably also affect their trade and lifestyle during the period of the story, and possibly the prosperity of the city, depending on whether said trade item is still in demand or has been made obsolete by some new invention or discovery.

There’s also the question of when and how those empires collapsed. Rot from within and barbarians from without, like Rome? War, leading to mutual exhaustion? Plague? Is any of that still a danger? And what’s left of them – a handful of more-or-less equal countries, or some small new places trying to expand into the decaying core of the original empire? I don’t plan on getting into lots of geopolitics in this story, but if my city is a trade center, what’s going on in it will be of interest to the rest of the world, and vice versa. Also, at least one of my characters is from away, so I’ll need to have her place-of-origin developed more than just “up north.”

Then there’s the city government (is it a charter city, like London, with a mayor and aldermen, but still answerable to the king? Or a city-state run by its own prince or council?), what the local factions are (besides my three kinds of mages), and a bunch of cultural stuff, especially cultural stuff revolving around clothes (because my main character is a seamstress).

Which brings in the question of what fabrics and decorations are available, and whether they’re produced locally (they don’t have to be; it’s a trade center, after all), which ones are expensive luxuries and which are the working-class wear, whether or not there are sumptuary laws. I know that unicorn leather is banned, but do they feel the same way about anything else?

I know there are at least some magical creatures in this world – fairies of the small-butterfly sort and unicorns, at minimum – so I need to know whether or not they’re intelligent and/or have their own magic, how different cultures treat them, and how the inevitable conflicts in attitude will get handled in this particular place. Possibly also how they’ll be handled in other places, if I end up with more characters who are From Away or who have traveled widely.

Many of these things, when I get them fully developed, won’t get into the story directly, but they’ll affect it profoundly because the historical and cultural cross-currents affect almost everything in the story. This is particularly frustrating for my story-noodler, because every time another bit of background clunks solidly into place, part of the plot changes, and she’s not used to it because I haven’t done this for the last eight or ten books. Also, because she doesn’t need to do as much of it, or not in advance anyway.

I’m not talking about writing specifics like dialog or characterization or syntax. I’m talking about process.

Let me explain.

Back when, I learned very quickly that if I took the attitude “I’ll fix it in the rewrite,” and simply plowed ahead as fast as I could without paying attention to the quality of my writing, my writing got sloppier and sloppier, until I wasn’t writing a first draft or even a zeroth draft, I was writing some semi-coherent notes that were hardly worth the time and energy spent on them. So I spent a lot of time learning not to get too far ahead of myself, and making it a habit to pay as much attention to how I was writing at any given moment as I paid to what I was writing.

Not that I give my Internal Editor free rein; far from it. But I found that some things were a lot easier to get right the first time than to hunt through the manuscript and fix later – things like unnecessary dialog tags, wordy or unclear sentences, descriptions that didn’t quite say what I wanted them to say.

Somewhere in the intervening thirty years, this way of working has…stopped working so well. Looking at it carefully, the problem appears to be that I got better at writing.

That sounds very odd, so let me unpack it a bit. The kinds of things I had to pay attention to, early on, were early-stage mistakes. As I got better at writing, getting those things right became a habit, and eventually almost automatic. Oh, my crit group still has to whop me upside the head every once in a while to remind me not to over/under-write, but it usually only takes one whop because it’s so obvious that all some has to do is say, “now, in this conversation here – ” and I interrupt with “Oh, rats, I did that thing again, didn’t I?” I also got enormously better at revising unsatisfactory stuff.

The combination means that slowly the kinds of things I need to pay attention to while writing changed. They got tinier and pickier on the sentence-by-sentence level, and larger and more sweeping on the structure level. A whole batch of new, not-possible-to-consider-until-a-draft-is-finished things cropped up in terms of plot flow and pacing and complications and balance, and I was still trying to get them right on the very first try.

The upshot is that, for quite a while now, there hasn’t been nearly as much payoff in paying attention to how I’m writing while I’m doing the writing. The things I most need to pay attention to have changed. I’m more interested in complicated plots and structures that require a lot of tinkering with after the first draft is done, because it’s impossible to tell on the first time through the manuscript what sorts of backfill will be needed and which scenes need to be added or deleted.

The other rather annoying change is that, due to the aforementioned complicated plots and structures, I need more pre-planning. My outlines are still all wrong, but they’re wrong in different ways from the way they used to be wrong. The characters are more likely to do what I thought they were going to do, but their reasons for doing it aren’t what I thought they would be, and this leads to needing more scenes on one side of the plot and fewer scenes on the other.

I’ve known since very early on that every book was at least slightly different in terms of the process it needed. Talking to Dragons and Sorcery and Cecelia were totally unplanned, sit-down-and-make-it-up stories; Snow White and Rose Red was much more constrained than usual both by the actual history I was playing off and by the fairy tale I was retelling; Mairelon the Magician had all sorts of charts of different characters’ relationships and position in the plot. Even so, I didn’t expect the particular change in process that’s crept up on me, especially since it’s a general change that appears to apply on a fundamental level to everything I write.

At least, it does right now. Ask me again in thirty years, and we’ll see how much else has changed.

He laughed and said blithely something along the lines of, “Oh, I thought you would talk about why you decided to use a strong female heroine (sic) in your first book Dealing with Dragons, and how you went about creating such a wonderful strong female-” for some reason, people who ask about this (and he’s not the only one by a long shot) always refer to Cimorene as a “strong female” and never as a woman or girl “-such a wonderful strong female. And if there’s time, maybe you could talk about the reasons you made her become the princess of a female dragon, and why you decided to make ‘King of the Dragons’ a title that has no connection with gender. That sort of thing.”

“I see,” I said. “All right.” And I hung up the phone. I was somewhat taken aback, partly because “female heroine” is redundant, but partly because I was in the rather rare position of being able to answer every one of the questions he brought up.

The problem was that every last one of his basic assumptions was wrong.

Dealing with Dragons was not only not my first book, it wasn’t even the first book I’d written in the Enchanted Forest chronicles. Talking to Dragons was published in 1985; Dealing came later, in 1990. And it is in Talking to Dragons that Cimorene and Kazul appear for the first time, as relatively minor characters. In that first appearance, Cimorene is not a “strong female heroine” – she is the main character’s mother, and she is exactly like all the mothers (mine and my friends’) that I remember having to deal with when I was sixteen. Well, maybe not exactly like them…

But in any case, the character who appeared later, in Dealing with Dragons, had to be someone who could reasonably be expected to grow up into the character in Talking to Dragons. Most of the things the moderator was asking about were my attempts to explain why Cimorene had turned out the way she did, not attempts to write about “a strong female heroine.”

It is also in Talking to Dragons that I first mentioned that “King of the Dragons” is a genderless job title (so far as dragons are concerned, anyway). I happen to remember very clearly writing the particular scene, because I was looking for a way to demonstrate that dragons were, in essence, aliens, not just very large human beings in lizard suits. I wanted a shortcut to show that they thought differently from humans, and I was very pleased when I came up with the idea of making the King of the Dragons a female. Similarly, Cimorene having been Kazul’s princess was an off-the-cuff invention that was intended to be a shock to the main character, the way it often is a bit shocking to a teenager to discover that his/her parents were once rowdy teenagers themselves.

In other words, pretty much all of the things the moderator wanted to know about were dictated by the needs of the first story and the personalities of the characters, not by the author’s desire to make a point in the second book.

This is something I think too many readers and would-be writers forget: most stories are not allegories, and the vast majority of characters in a non-allegorical, realistic piece of fiction are not going to work if they are portrayed first as a member of a group (“a typical ___”), and only second as an individual with whatever strengths and weaknesses, quirks and phobias, that particular individual happens to have.

And I would argue that regardless of what traits or attributes a character has – race, size, ethnicity, sex, age, hair color, etc. – what shapes them most is the interaction between their own personality and the attitude of the culture they grow up in toward their particular traits, because that pretty much determines both the way the characters think of themselves (and others) and the ways they expect other people to think and behave.

Dragon culture, as it developed in the Enchanted Forest books, pretty much ignores gender because dragons get to choose which sex they’re going to be. It’s important, but it’s important to a dragon in the same way as choosing the right hairstyle is important to human beings. They wouldn’t judge another dragon’s intelligence, competence, or abilities based on what gender that dragon was. Kazul is a fairly normal, mainstream member of that culture, so she doesn’t think twice about being king, and she’s perfectly happy having her princess learn magic and carry a sword.

Cimorene, on the other hand, grew up in the fairy-tale-kingdom culture…and rebelled against it. That’s what I mean when I say it’s the interaction between the character and the culture that really makes them who they are in the story. If I had written Dealing with Dragons first, and I’d reversed the personalities and made Cimorene a normal, mainstream princess and Kazul a rebel against dragon culture, they’d have been very different characters…and I could have chosen to do that. Which would, of course, have led to a very different story, but that’s another matter.

Recently someone asked me what I meant. Surely, if you want to write, that kind of assumes you want to be a writer.

Well, that depends on how you define “writer.” If you define “writer” as someone who writes, then “being a writer” is trivial: you sit down and write something, and presto! You’re a writer. It’s not this huge thing to aspire to, because it’s too easy. By that definition, I’ve been a writer since I started working on my first story back in seventh grade.

Most folks, of course, aren’t talking about that basic definition when they speak of “being a writer.” What they mean is being a published writer, a career writer, a professional writer, a full-time writer. And what they’re really asking about is rather more complex than the question sounds.

Because there are two parts to “being a writer” the way they mean it. There’s writing, and there’s what happens afterward. Most of the people who talk about “being a writer,” whether they’re asking the question or whether they’re announcing to the world that they themselves want to be writers, are either talking about the perceived glamour and respect and status that they think goes along with publication and a writing career, or else they’re talking about the validation of getting published – the fact that someone, somewhere, has deemed their story worthy.

All that is stuff that happens after the writing part, if it happens at all. (And when it does happen, it’s nothing at all like the rosy dreams people have of what it’s going to be like…but that’s a different post, I think.)

Wanting to write means wanting to get the words down on paper (or, these days, pixels); wanting to tell stories; wanting to get the stories right on as many levels as possible. It’s not about the stuff that happens to the paper or pixels after the story is written.

Which is not to say that I never desired publication; on the contrary, getting my stories published was a goal from the time I realized such a thing might be possible, which was around age thirteen. But I never wrote in order to get published. I wrote in order to get the story down and get it right. Publication was one of several possible proofs that I’d done what I set out to do; it was also the most effective way of getting the stories out and read by other people. (This was not only pre-Internet, it was pre-personal-computers.)

I don’t remember publication ever being the same kind of goal that the writing itself was. I didn’t sneak time in class or jot weird notes in the margins of my textbooks because I wanted to be published. I didn’t spend my lunch hours and coffee breaks at my office typing instead of chatting with my coworkers or eating with my friends in order to get published. I did it because I wanted to tell the story. I wanted to find out how I would get it to turn out right. (Sometimes, I just wanted to find out what would happen; I don’t always know in advance.)

People who talk about “being a writer” usually don’t want to write; they want to have written. They want to skip ahead to the part that comes after the writing. They want the status, or they want the validation.

And while validation is lovely, and all the publication and publicity stuff is certainly a necessary and legitimate part of a writing career, they’re not writing. Hardly anyone who wants to have written makes it to publication, and if they do, they usually don’t continue past one or two books, because if you are going to “be a writer,” you spend 80% of your career time writing. Not doing the stuff that comes afterwards. And 80% of your time is way too much to spend doing something you don’t really enjoy, just to get to the “good parts.”

The flip side of this, of course, is that unless one aspires to be Emily Dickenson and only ever publish posthumously, one does have to think about selling and publication and publicity at some point. Validation is important; so are sales (especially if one hopes to make a living at this). Ignoring or sneering at the business end of writing is just as problematic as wanting to skip over the writing part and get to the afterwards.

The difference is that if one truly can’t stand the sales, publication, and publicity part, one can skip it completely. Nobody goes around arm-twisting people into sending their manuscripts out, or querying editors, or even just putting their stuff up on one of the freebie web sites. If one wants to be Emily Dickenson, nobody will stop you (though people will probably look at you funny if you tell them, so perhaps it’s better not to mention it).

If, however, it’s the writing part that one strongly dislikes, one is pretty much up a creek. You can’t sell a book that hasn’t been written yet (not the first time, anyway…and it’s getting harder even for writers with a proven track record). You can’t publish a non-existent manuscript, or even an incomplete one; you certainly can’t do all the sales-and-marketing stuff for a book that doesn’t exist.

What it comes down to is being honest with oneself about why one is doing this. The professonal writers I know range from barely tolerating the publication-and-after stuff to reveling in it with great glee, but what keeps all of us at this job is the writing part. Telling stories. Making things up. Even for the writers who most enjoy the publicity bits.

I got better at it by practicing, but it was a long, slow process because I still wasn’t thinking explicitly about character – what it is, how it’s expressed, how the reader learns about it. It wasn’t until many years and many books later, when I had to explain to someone how it worked, that I started to see what I was doing and figure out how to do it better.

There were several layers of realization for me. The first one was kind of a “duh!” thing – you find out about characters the same way you find out about actual people. You figure out what they’re like based on their physical appearance, what they say and how they say it, what they do, how they act and react, and on what other people say to them and about them, and how people you already know react to them. If someone you already like and trust says that George is a good guy, if a little stuffy, you’re inclined to believe him and give George the benefit of the doubt; if someone you distrust and think is a bad guy says that George is a perfect example of honorable behavior, you’re a lot more inclined to count your change twice when George hands it back to you than to trust him with your wallet and I.D.

The next realization came when I figured out why so many other people thought I didn’t have much problem with characterization – it was because I did dialog reasonably well, and that meant that readers were able to judge my characters based on what they said and on what other characters said to or about them. I was leaving out a lot of the other things that would make my characters deeper and better rounded, but the dialog was enough for a lot of readers to go on with, especially in the sort of adventure novels I was writing at that point.

At that point, I started paying more attention to some of the things I’d been doing by instinct; that is, I started trying to deliberately educate my instincts so that they’d work better without me having to constantly watch what I was doing. Mostly, this involved thinking about exactly how one presents all those aspects of a character’s personality in a novel.

Physical appearance looks easy, at first glance. It’s just a description of what the person looks like, right? Well, yes. But physical appearance is more than height, weight, and the color of hair, eyes, and skin. It includes clothes, which in nearly every society in history have been a marker of class, status, and degree of general coolness, and often of occupation and/or education as well. Things like cut and color, fit, fabric, style, whether clothes look/are comfortable, the degree of repair they’re in, how becoming they are to the person – mentioning just one or two of these can tell a lot about a character and his/her situation. The same goes for a character’s hairstyle and, in the case of men, whether they wear a mustache and/or beard, in what style.

What the character says and how he says it covers tone of voice and vocabulary as well as syntax. It includes things like whether the character is very blunt (“No. Not ever. Not for a million dollars.”), less blunt (“I won’t be available Tuesday. Or any other day.”), or vague and non-committal (“I’ll just have to see how things go.”); whether she yells or whispers; whether he’s gentle or sarcastic or abrupt or abstracted.

What the character does and how she acts divides into two parts: body language and actual actions. Facial expression – smiles, frowns, narrowed eyes, raised eyebrows, twitching lips, blushes – get included her, but so does every other part of the body, which some authors tend to forget. Things like stiffening, turning away, crossing one leg over the other, waving a hand, leaning forward – all these are part of a nearly unconscious mode of communication that all of us do in real life all the time. It’s so nearly unconscious, in fact, that many people have to go to some lengths in order to start seeing it so that they can describe it piece by piece, because they don’t naturally break down “He was interested” into “He leaned forward, eyes fixed on the contract, lips pursed slightly as if to keep from admitting anything too soon.”

For me, my characters’ body language ends up being sort of like method acting. I’ll be writing a scene and type “He was interested” and immediately know I want the specifics. So I act the character in my head:  I’m him; I’m interested; what, exactly, is my body trying to do? Oh, I’m leaning forward – hands are twitching a bit, but they’re under the desk, so nobody would see – eyes want to squint – what’s my mouth doing? Shoulders?   Once I figure all that out, I decide which bits to put into the description, and opt for the pursed lips rather than the tense shoulders.

The other part of “what the character does” is action, which is more movement than body language and comes in two subtypes. First, there’s immediate short-term behavior – whistling, slamming a door, crying, laughing, slapping someone; second, there are more complicated, long-term actions, like buying someone a gift, running away, challenging someone to a duel, pretty much any sort of plot-related activity. With immediate action, I find that the context and body language part is as or more important than the action itself. Someone who flounces out the door, slamming it behind her, does not come across as a threat, while someone who storms out and slams it hard enough to break the glass is a lot scarier.

The hardest part is often figuring out what that particular character would do in a given situation. Would he fidget with his pocket watch? Hum softly? Pace? Start studying the view out the windows? Sigh, softly or noisily? Unobtrusively finger the dagger up his sleeve, in a way that makes everyone watching think that he’s fluffing the ruffled cuff of his shirt?

How the character reacts to different people and situations builds on everything else, because the way they show their reactions is in their dialog and tone of voice, their body language, their immediate and longer-term actions. You can, of course, simply say “Carol disliked Jane instantly,” but it’s usually much more effective to say “Carol stiffened more and more as Jane simpered through George’s rambling introduction. When George finally finished, Carol inclined her head a quarter of an inch. “Pleased to meet you,” she said in an icy tone.”

The longer version is more effective partly because the readers can judge Carol’s reaction for themselves, but also because George’s and Jane’s actions hint at what they’re like and why Carol might be having the reaction she’s having.

I still do most of my characterization by instinct – that is, I don’t get this analytical when I’m actually writing a scene. As I said, for me, it’s more like method acting – trying to be the character for an instant or two, long enough to figure out what to describe. But taking it apart this way helps me educate my instincts, so that I don’t have to stop every time one character is introduced to another whom they dislike, and make lists of all the possible ways she might show her reaction, then consciously and deliberately pick out the one thing that would be right for that character to do/say/think. If I do my thinking about the mechanics of how characterization works outside my actual writing time, I don’t have to do it when I’m trying to figure out the scene.

But that’s me. Your mileage may vary.

I didn’t. I never, ever wanted to “be a writer.” I wanted to write. I wanted to tell stories. I wanted to get these blasted characters out of my head and nailed down on paper so I wouldn’t have to keep thinking about them.

Being a writer is something that happened as a result of writing, almost by accident. It was never my goal. My goal was always to finish the current story, and then come up with something even cooler to write about next time. Publishing and making a living were afterthoughts.

At what point in your life did you think you could actually make a living from your writing?

About five and a half books in. That is, I had written and sold five novels, of which two (I think) were somewhere in the production process, and I was partway through the next book, which I was about to send off to my agent to sell. My first book had earned out by then, and I think the second had, too, so I had variable royalty income from those, plus the known amounts I was getting as the second- or third- partial advance payments on the two that were in the production process. This meant I had a pretty good idea what my writing income was likely to be over the next year or two.

At that point, I’d been thinking about quitting my day job for a few years, so I’d been building up a savings account in anticipation. The idea was that I’d have enough cash to get me through a dry period or two, and if it ever dropped below six months’ living expenses, I’d start looking for a new day job (figuring that six months would be long enough to find one). I’ve had to dip into that fund several times over the years, but it’s never gone below the six months line (knock wood).

So if you’re asking when I started thinking about quitting the day job (and planning and preparing to do so), the answer is some time around 1983, roughly two years before I actually quit and went full-time. It didn’t become a serious possibility until I had the income and the bank account in place, which took two years to get fully set up.

When you work with fantasy, how is it different from something like realistic fiction?

I wouldn’t know; I’ve never written anything that wasn’t fantasy. I did try once, but one of the characters turned out to be a wizard in Chapter Two, and I gave up.

Still, I think I can say a little more than that. The basics of writing are the same, regardless of genre: style, viewpoint, dialog, characterization, plot, etc. Sometimes there are genre conventions that are important and that can expand or limit the range of techniques that are available to the writer in that genre, but by and large, effective writing is effective regardless of content.

Worldbuilding and background tend, I think, to be a bit more important in speculative fiction in general than in so-called realistic fiction, simply because one can choose to set realistic fiction in places that the reader is likely to be familiar with already, and which therefore need much less development in the story. That’s about all I can think of, though — and it’s not a hard and fast rule. Lots of realistic-fiction authors set their novels in places that their likely readers will consider exotic (whether that means New Orleans or Tokyo, Los Angeles or Paris, Moscow or Sydney). Part of the point of doing so is to give the readers a chance to image places that are strange to them, which requires just the same sort of in-story background setting as any SF story.

What are some of the criteria you look at when first starting a piece?

Sometimes, there a business considerations that dictate what comes next; for instance, if I’ve signed a contract to write a trilogy, then when I finish the first book, the next piece is going to be Book #2. Or if my agent is trying to re-sell some of my out-of-print backlist, sometimes it is easier if I promise to write a new sequel. Or I need to write something to fulfill the option clause in a contract before I go on to what I really want to write next.

But apart from a couple of multi-book contracts, business considerations haven’t come up terribly often for me, so the main thing I think about when I’m deciding what to write next is, “Is this a story I’m interested in writing?” Since I usually have anywhere from three to twenty possible stories for which the answer to that question is “yes,” the next question is “Is this story insisting on being written now?” If one of them is, then that’s the one that comes next. Usually, there isn’t any one piece on the list that’s at critical mass and/or chomping at the bit to get going, so the next questions are “Which story(s) are almost ready to move forward and/or can be gotten to that point with the least amount of work? Or which one(s) will be the most fun to play with, even if they’re going to be a whole lot of work to get moving?” and “Of the stories that appeal to me and that I think I can get moving, which one(s) does my agent think she can sell most easily in the current market?”

That usually whittles the list down to one or two titles, at most, and if I still can’t decide, I flip a coin.

The other big advantage they have is that they don’t seem to get as many fans asking about obscure worldbuilding points, some of which aren’t even in the story. I’ve never heard of someone coming up to a writer who has written a series of historical novels set in New York City during the American Revolution and asking “So, I’ve been wondering what was happening in Australia while all this was going on.” And if somebody did ask, I know of nobody who would think the writer out of line if he answered, “How should I know? Google is your friend…”

But when you’ve invented a world, readers do this sort of thing all the time. I still remember the fan letter I got from a gentleman who’d read The Seven Towers that went something like this:

“Dear Ms. Wrede: In your book, you mention the Three Greater Obligations and the Twelve Lesser Obligations. I can only find nine Lesser Obligations in the text. What are the other three? Sincerely yours,”

For those of you who haven’t read the book, the greater and lesser obligations were part of the culture of a secondary character, a foreigner who was the only member of his group who ever came onstage during the story (though we heard a lot about them). Since only the one character was actually in the book, I didn’t bother making up the culture in detail; when he brought up the Three Greater Obligations, I knew what they were because they were important to the situation, but when he mentioned the Twelve Lesser Obligations, I figured that was enough to cover anything that was likely to come up in the course of the book, and I didn’t actually need to have a list.

So when I got that fan letter, I didn’t have an answer. Which tends to surprise and annoy the sort of fan who so earnestly asks questions like that. For some reason, they’re positive that I have several sets of virtual encyclopedias, one for each of the imaginary worlds I’ve created, that cover everything anyone could possibly want to know about their history, geography, cultures, magic, and so on.

It doesn’t work that way for most of us. Yes, every so often you get a curve-wrecker like J.R.R. Tolkein who spent forty years inventing everything from languages to poetry for his imaginary world – but those people are nearly always doing it for fun. As a hobby. Because they like making up every possible detail of their imaginary world.

Most working writers don’t have that much time, not when we’re trying to make a living as writers rather than Oxford Dons, and especially not when we’re working with multiple different imaginary places. What we do instead is what I call the soap-bubble technique – we know a small number of key details, the sort that imply a lot of other interesting possibilities, and we scatter them through the story instead of giving them all to the reader at once. Like taking a drop of soapy water and blowing it full of air, this gives the illusion of a sizeable object much larger than the actual material that makes it up. There isn’t anything in the middle but air, but it doesn’t matter because the bubble is so pretty and it doesn’t actually have to last any longer than the story it’s background for.

Furthermore, some of the best and most important details in my books turn out to be things I made up on the fly. The interesting contradiction here is that I need to have put considerable thought into the background before I’m able to do that sort of on-the-fly invention…but most of it doesn’t have to be at the detail level. I need a structure that things have to fit into, so that everything I come up with stays consistent, but I don’t need all twelve of the Lesser Obligations, especially when I don’t plan on mentioning any of them specifically in the text.

Sometimes I do work out unnecessary extras, just for fun. When I was writing The Raven Ring, I worked out the entire fortunetelling deck of cards and their meanings, just because, even though I only needed ten or so cards in the actual text. I had an obscure secret history behind them, too, though none of it ever got into that book. But that was just because I was having fun, not because I had to know all that in order to write the book.

There’s one more factor involved in not-making-things-up besides the time and energy: the problem of being trapped, of needing something to be X in order for the plot to work, but it can’t be X because you’ve already made up Y. Not “you’ve already put Y in the book.” If the background gets too full of specific, interlocking, irrelevant detail, it can cripple one’s ability to suddenly see a completely different possibility…because the new thing isn’t a possibility; that part of the background is already filled in.

It’s a delicate balancing act. Every writer has a different threshold for how much detail is enough, how much is too much, how much has to be done in advance, how much can be made up as needed. Sometimes it changes from book to book. The point is, the threshold can change, because all a fantasy writer really has to worry about is internal consistency. True, most of us set our stories in worlds that have some vague connection with reality – that have horses and rabbits and laws of physics that are mostly like ours (except for the magic part). Where there’s overlap, one does research. But there’s always the possibility of something different – there don’t have to be horses or rabbits or the laws of physics as we know them.

And possibility is, for me, what writing in general and fantasy in particular are all about.

No writer I know works this way, not even the ones who really do pick audiences and make outlines.

I’ve been thinking about this a bunch lately, because I just finished a book and I’m in the process of booting up the next one. And it occurs to me that the very first thing I do is decide why.

Why covers a lot of things: why write at all, why start another book when I have so much else going on, why pick this book to do next instead of that one. There are a lot of answers, but the one answer that it occurs to me I have never heard from other writers is “to get published.”

Now, possibly this is because publication is a milestone that most of my writer friends have already passed, but I don’t think so. For one thing, selling one story is no guarantee that you’ll sell the next. For another, I don’t hear it from my unpublished writer friends, either. Not if you ask them “Why are you writing that story?” Answers range from “For fun” to “I just have to,” but “To get published” is never what anyone comes up with first. Publication is always tacked on at the end “…and of course, I’d like to get it published one day.”

I mentioned this to Beth-my-walking-buddy and she pointed out that publication is the validation, not the motivation. It’s the thing that says I did a good job, not the reason I’m trying to do the job in the first place.

“Why am I writing this?” is not actually something I think about all that often, but knowing whether I’m writing a story to fulfill the option clause in my contract, to make one of my friends smile, because the idea wouldn’t leave me alone, because I have bills to pay, or because this is a story I am desperately in love with and want to tell, does make a difference. Sometimes more than one thing is true at a time, and it’s easy to forget that wanting to tell the story is really more important to me than paying the bills. And when I forget why I’m writing and what my original vision of the story was (the one that got me excited about it in the first place), I tend to wander off track, and eventually things bog down and get difficult.

The second thing I do when I’m booting up a new book is brainstorming. Sometimes, it’s just tossing ideas around in my head; sometimes, it’s the kind beloved of corporate managers, where I sit down with pen and paper and draw spidery diagrams all over a page; sometimes, it’s focused on one particular aspect of the story. At the moment, I have two of these going: the first is an untidy heap of ideas, everything from scraps of possible dialog to potential characters and backgrounds to plots to “things I would like to see happen” (Max chewing out Jillian, for instance). Some of these will end up in the story, some not.

The other is a focused brainstorm on sevens – that is, lists of seven things (seven deadly sins, seven cardinal virtues, seven chakras, seven colors of the rainbow, seven holy mountains, seven wise men, seven wonders of the world, seven habits of successful people…every kind of seven I can think of or find). This one is because I know that my main character will be facing seven related tasks or tests, and noodling around with all the other sevens people have come up with makes me look at lots more possibilities for how to link my tasks together. I don’t actually plan on using any of the real-life lists as the basis for whatever I come up with; they just sort of get me in the mood, and then I start making my own lists of seven things that might go together the way I want them to.

Eventually, I’ll have enough of this story-stuff heaped up, and I’ll organize it into a plot outline (the third step), and then I’ll buckle down to serious writing. The point is, the outline comes rather far down the process (brainstorming for a whole novel can take a while). Outlining is not even a requirement; it’s just a tool for organizing all that brainstorming that I find useful.

I think that all writers go through this sequence, though few of us break the process down into steps (and some of the steps moosh together, or happen so fast that the writer doesn’t even notice). For those who don’t bother outlining, the organizing and writing happen together; for writers who write to find out what happens next, the brainstorming and the organizing and the writing all happen at the same time; for of the “sit down in front of a blank screen and surprise myself” variety, even the vision of what the story is and could become happens as the words go down on the page one after another. And there’s no particular reason to slow down and try to do the parts of the process one at a time, unless the just-sit-down-and-write thing stops working for a while.

I do think that it’s useful to think about this stuff, because it allows me to notice when I’m trying to do things in the wrong order. If I think of my outline as a necessary first step, instead of as a tool for organizing all the brainstorming, I get extremely frustrated when it doesn’t go well. But really, if I haven’t done the brainstorming, if the story-stuff hasn’t reached critical mass, there’s nothing to organize. And a generic outline (“There are some good guys who have a problem. They start trying to solve it, but they have trouble with some bad guys…”) is pretty useless.

Which means I am now looking at my huge list of Possible Things To Write and contemplating which idea(s) to start poking at. My agent has weighed in, and so have several of my friends; they’re all pretty much in agreement, so unless my publisher gets really demanding about some other possibility (and does so pretty soon, before I’m totally committed to this project), I probably have settled on The Next Thing.

And what it started with was this:

No shit, there I was –

What, you don’t like the opening?  Listen, it’s fairy tales that start “once upon a time.”  War stories are supposed to start “No shit, there I was.”

So, no shit, there I was, thread in one hand, needle in the other, and a silk bolt worth four thousand isiri spread over my lap, when -

Now what?  Oh, you think this doesn’t sound much like a war story?

In the last few weeks of thinking about this rather minimal story-seed, I added a McGuffin (although I have no idea yet why it’s significant), a notion of what happens in the first half of the opening scene, and the barest hint of a plot thread. Oh, and two, count them, two secondary characters, one of whom probably won’t be around for more than two chapters, tops.

This is not much to start writing a novel with.

I could just take what I have and keep writing for a while, to see what happens and what I come up with. I already know, however, that this seldom works well for me, so I’m not going to start by trying that. I need to develop what I have a bit more, until it gets past the Critical Mass point and really starts rolling, and that means poking at what I have until new things show up and start to gel.

The question always is, where and how to poke. Up until last weekend, the obvious point to poke at this story was the characters. The story needs more of them, and I need to know more about the few that I already have (well, about two of them, anyway. I don’t think I really need to know much more about the one who’s disappearing within two or three chapters). And characters and what they want or need (but can’t have…yet) are the heart of most stories.

So I’ve been thinking about these people off and on: who they are, where they come from, what they’re each trying to do and why. I was thinking about the second character, the one who’s not the protagonist but who will be a major player, and why that was happening…and I figured out something about the McGuffin. And suddenly, I had a structure for my plot.

As soon as this happened, where I need to poke at this idea changed. See, structure is fundamental for me. It’s what goes under the plot, to hold it up. What I need to know next, for me to be able to finish that first scene, is what I’m going to build on that structure and why. Once I know that, I’ll know who the rest of the characters have to be and what they’ll have to do. Undoubtedly, that will change the plot – once I have characters and they start acting and interacting, they always end up changing the plot details. That’s what makes it all work, for me.

But the characters and incidents won’t change the structure. That’s solid. I know how many incidents I need, and the effect they have on the McGuffin; now I need to figure out what they are and why the villain set things up this way and how they’re going to affect my characters. (I’m not too worried about how my heroine is going to mess up the villain’s plans; after that opening, I have no doubt she’ll think of something.) Oh, and I need a villain…the structure requires one.

If this were going to be a different book, or if it had started with a different set of bits – say, a well-developed setting and a bunch of characters, but no plot or structure – I’d probably have started by poking at the characters. The point isn’t how I’m doing this, or that anyone else ought to work the same way. The point here is: 1) The basic idea needs a lot more development before I can make much forward progress; 2) The development doesn’t just happen; it requires poking; 3) Where I poke keeps changing, depending on how much I’ve already figured out.

Changing where I poke at ideas is part of the process of developing them. I don’t make up a list of characters, then figure out everything about their backgrounds and personalities and desires before I ever start thinking about plot or setting. I think about a character for a bit, then about the McGuffin for a bit, then about a different character, then maybe about the setting/history/culture.

This morning, in conversation with Beth-my-walking-buddy, I got a handle on the villain, and the whole plot changed. So did one of my two supposedly-known secondary characters. The structure’s still the same, though, and so’s the McGuffin; a little more background, and I’ll be ready to start writing my first totally-wrong outline.

(Julie D, I’ll put up the post on agents on Sunday, when I’ve had a chance to think about it a bit more.)

I am – finally! – within three chapters of the end. (I think it’s only two more chapters, but things often take longer than I expect them to, so I’m allowing myself three just in case.) And I am therefore confidently expecting (absent any additional crises) to have the whole thing done by the end of the week.

Yes, that means I expect to write two to three chapters in four days. No, this is not my usual work speed. So why am I so confident?

Barn door syndrome. Like the horse coming home at the end of a long day, I can see that I’m almost there, and no matter how tired and sick of this book I am, the thought of being done provides all the oomph I need. Also, with only 2-3 chapters left, I have a very clear idea of What Happens Next, and very little time for unexpected twists to mess things up. The next scene will be the Last Big Crisis, followed by whatever immediate clean-up is needed; that should complete the next chapter, or a bit more if handling the Big Crisis takes more space than I think it’s going to. Then I have some character issues to get finally resolved, (half a chapter, or possibly a whole one…characters will keep talking on, even when I really want them to get on with it). Then I can send them all home at last.

I’ve been anticipating this rush-to-the-end for the last ten chapters. I was really hoping it would hit about five chapters ago, but that was before I realized how much I still had to cover in detail and how much space it would take.

Which goes to show that one ought not to depend too much on one’s previous process or productivity levels, as they can and do change without notice. I am already having to tell myself firmly, on an hourly basis, that just because I am currently producing nearly a chapter a day, this does not mean I can blithely assume that I will continue at this rate, and can therefore figure on writing my next book in a month or less (much as I would like to).

On the contrary, experience shows that my daily word-count production is likely to drop way, way back as I fiddle with plot outlines and plans. It’ll spike for a couple of chapters around chapter two or three, then I’ll hit a wall somewhere between chapter 7 and chapter 10. There will be a long, slow slog (punctuated by quicksand and distracting emergencies) until somewhere around chapter 20, where things will start picking up until I once again get within three to five chapters of The End, whereupon the rest of the book will come out in a rush.

The pattern is fairly reliable. The trouble is figuring out how many chapters the book will have (and/or how long the chapters will be, as a shift there can throw off the pattern quite easily). I was expecting this book to be around 25 chapters. I finished chapter 31 this morning, and have, as I said, two or three to go.

You’d think I’d know better than to count on the predictabilty of my process by this time.