The most obvious and least practical method is to write the entire epic before allowing any of it to be published. This has the advantage of treating the multi-volume series the exact same way as one would treat a story complete in one volume: you write the whole thing, you edit and revise the whole thing, you review the whole thing for consistency and pacing, and then you finally publish it. Unfortunately, very few writers are in a position to do this with even a short series or trilogy (not to mention that most of us lack the patience necessary to do without readers for so long), which means that some of the books will probably be in print and un-revisable before the end of the series is even in first draft.

That leaves the would-be epic novelist with one option: prevention. It’s not foolproof, but it’s better than ignoring the whole issue.

The first thing to do is to understand the pitfalls. Really understand them, not just as a check-off list, but as things you can recognize almost as soon as you see them. If you don’t recognize something as problematic, it’s almost impossible to fix even after the fact; preventing it from happening in the first place will likely be a matter of luck, no more. Also, there is no writing technique that is always a bad idea. If you understand the potential problems, then you’re also more likely to understand when they aren’t problems, and when adding another viewpoint or subplot or volume is a plus for your story, rather than a minus.

The next thing to remember is that prevention involves a certain amount of planning ahead. This can be tricky for the sort of writer for whom outlining or telling the story kills it dead, but it’s usually not completely impossible if they avoid the particular areas (usually plot) that do the story-killing, and focus instead on more abstract aspects of the story.

For the rest of us, planning ahead usually begins with some kind of shape or structure. The Harry Potter series is shaped by the British school system; each book covers one school year until Harry reaches what should be his graduation year. That shape or structure is more or less inherent to the story Rowling was telling, but many stories don’t have such a tidy shape embedded in them from the beginning. For those, the author has to find or choose or invent the framework that will support the story: the seven deadly sins or cardinal virtues, one per book for seven books, for instance, or an invented set of tasks to be covered, events to happen, places to go, or people to meet.

While this kind of shape or structure, or even an arbitrary limit on the number of volumes, can do a lot to help an author keep a story under control, it isn’t absolutely necessary. The important thing is the control, not the specific mechanism by which it is achieved. The longer the journey, the more necessary it is to have a road map and compass, and to check them frequently to make sure one is still on track. (Sadly, they have not yet invented a GPS for writers that will break in while you’re typing to say “This scene is off track; your characters will never get across the mountains this way. To get back on track, delete the snow-elves, mystic polar bears, and cloud-fairies and have your characters go down Caradhras and head south to the Mines of Moria instead.”)

Then comes the outline, which is only for people who actually do outlines. People who can’t outline or who go by instinct still need something, but it’s usually not specific incidents or a plot line; it’s more of a feel for “what this story is.” Whatever you do, you will probably need it to be clearer and more detailed than you think, because the basics of prevention involve regular checking of what you write against your outline or feeling. It’s not a matter of rigidly following the plan; you’re allowed to decide that I-70 from Denver is closed, so you’ll have to swing south through Utah to get to L.A. What you’re trying to do is make sure that you don’t end up in Mexico City while you’re still promising everyone that you’re going to get to Los Angeles one of these days, yes, indeed you are.

To do this, you establish a routine of checking back with your road map/outline every ten chapters, or every 25,000 words, or every third-of-a-book, to make sure that what you’re doing is still heading in the right direction. If it isn’t, you then need to decide whether you can keep your shiny new characters/subplot/background and get back on track without too much of a detour. If it isn’t possible, you grit your teeth and take it out. The idea is to set your check-in so that it’s frequent enough that you won’t end up trashing half a book or more, but not so frequent that you start feeling like it’s a straightjacket that takes all the fun out of it.

Checking in at the end of the first draft of each book is not optional. This is where the prevention part really kicks into gear, because whether you’re following an outline or writing an epic by the seat of your pants, you are going to be stuck with whatever you’ve written for the rest of the series, so you need to make sure you can live with it. If you are particularly methodical, you can, at the end of each book, make a list of all the viewpoint characters and how many scenes they each have, or do a chart of all your subplots and where they are and where you expect them to go. The idea being, of course, to see if they’ve started proliferating madly on you, so you can catch them while you can still do something about it.

At this point – the end of the first draft for Volume-Whatever in the series – you have a choice: you can either revise backward, or revise forward. That is, you have your middle-of-series draft, which has started developing in unexpected directions. You can either trim it back ruthlessly so as to keep to your original vision (backward revision, i.e., revising the book you have just written), or you can change your vision of the story (forward revision, i.e., revising your outline or concept or whatever you’ve been using to keep things on track).

Be aware that revising your outline/concept is a lot of work. If you’re still early in your epic series, say book 1-2, you can rip up and rearrange major plot threads without it being too noticeable, but this will mean essentially redeveloping the entire rest of the series plot outline. The farther you’ve gotten in your epic, the harder it is to change course. This is possibly one of the reasons for Epic Bloat – if the writer has a Cool New Idea halfway through Book 3 of a six-volume epic, it looks like being easier to add another three or four volumes to the series (thus making the change barely 1/3 of the way through, rather than halfway through). Don’t. Really. It will not end well.

To sum up: preventing Epic Bloat is mostly a matter of paying attention and being ruthlessly honest about what is and isn’t necessary to the story. To do this, you have to know what your overall story is and how your subplots fit into it, and you have to keep checking as you write and finish various volumes to see if you are still writing the story you set out to write.

Her first and biggest point is: Pick a structure (a specific number of books) and stick to it. The difficulty here is that for a lot of epic series, this is essentially an arbitrary decision…and the author knows it is arbitrary. It’s one thing when the story itself falls into obvious, well-defined chunks (as with the Harry Potter series, where each book covers one school year). It’s another thing entirely when one sets out to write an epic of, say, half a million words that doesn’t fall naturally into neat one-to-two-hundred-thousand-word segments.

You can, of course, pick a number and figure you’ll cover the story in that many books, period. That can cause other problems down the line, though. If you’ve guessed right, it’s not a problem, but if you’ve over- or under-estimated the amount of story you have, and don’t discover this until mid-series, it can be impossible to stick to your resolve without harming the story.

It’s fairly common for authors to mis-estimate the length of a novel by ten to thirty percent, sometimes more. When you’re looking at a single, hundred-thousand-word novel, that’s an extra 30,000 words, which is often OK with the publisher (and if it isn’t, it’s usually possible to edit the book down to within shouting distance of the publisher’s limits). When you’re looking at a 500,000-word, four-to-five-book series, that’s a swing of 50,000 to 150,000 words, or up to one entire additional novel…and by the time one realizes that, the first two books are usually already on the shelves and cannot be edited. Making all the length adjustments in the last two books of the story can be next to impossible; adding one more book is a lot simpler, especially if it’s a popular series.

And once the author has added one book, it’s easier to do it again. Absent an obvious natural structure like the number of school years, there’s no real reason not to add one more book…and that opens the door for a lot of the other problems Ms. Brennan identifies in her post, the next of which is control your points of view.

Modern epic fantasy seems to be written mainly in tight-third-person with multiple viewpoints. One of the reasons for this is an extreme addiction to “showing” things rather than “telling” them. This manifests not merely as a reluctance to cover offstage plot points in narrative summary, but as a reluctance to allow the characters within the story to tell each other anything. If the messenger or dying villager isn’t allowed to tell your POV character about the burning of the village, your only alternative is to “show” it by presenting the scene in its fully dramatized glory…which generally means adding a POV character, because none of the main characters or existing POVs were around.

I blame this addiction to dramatizing scenes (and the consequent multiplying of POV characters) squarely on movies and TV. There’s good reason for it on screen; time is limited, and it’s a lot more efficient and dramatic to show a five-second shot of a hotel blowing up than to spend thirty seconds on the scene where the messenger describes it. On-screen, you also don’t have the same viewpoint problem – the camera is the viewpoint, whichever character it chooses to follow. In a novel, it works the other way: narrative is usually more compact than dramatization, and you do have the viewpoint problems.

Adding a viewpoint can also be an easy way to get offstage information to the reader; it can also the reader know more about what is going on (if they can keep track of it) than the main characters do, which supposedly makes the readers feel clever. Sometimes there are Really Cool Bits that the writer simply can’t put in unless somebody is actually viewing the scene (“murder your darlings” anyone?). Early in the series, the writer may want to foreshadow something or establish characters for later, and by the time they turn out to be unnecessary, the first book is in print and it’s too late to cut the scene. And finally, POV characters proliferate on occasion because the writer likes them and really, really wants to write their POV.

This leads directly to Ms. Brennan’s next points, which are control your subplots and centralize. The trouble with introducing a new POV character is that every person is the hero and central focus of his or her own story. The minute you give a character a tight-third viewpoint scene, that character starts reconceiving the whole story on their terms, and brings in all their personal concerns about their gambling debts or their son’s education. And unless they’re a throwaway viewpoint, like the villager who gets killed in the raid just so the writer can “show” the raid onstage, the writer is very likely to have to deal with some of those concerns, which means more scenes for that character, as well as a whole new branching tree of subplots.

The second reason for proliferating subplots is the problem of balance. If you have two viewpoint characters who are both supposed to be central to the story, you really want to give them almost-equal time on stage. This means that if you have one character who is sitting in town having several chapters-worth of adventures in a couple of days, you don’t want your other viewpoint character to be having a boring two-week voyage from point A to point B. So you either have to stretch the timing of the in-town adventures (which can be tough to make plausible), or you have to have your traveling character run into some interesting trouble so as to keep the timing and emphasis the same, or you do what Tolkien did and spend several months and many chapters on one character, and then abandon them and spend an equal number of time and chapters on the other character(s).

The other big reason writers fall into this kind of trap is that a multi-volume story feels, initially, as if it has loads of room for all this stuff. The whole point of an epic is to be able to spread out and dig down into the detail, isn’t it? So it’s easy to throw in lots of subplots at the start, without quite realizing just what it’s going to take to develop them and then bring them all to a satisfactory conclusion.

The crowning problem is the one I mentioned earlier: the problem of publishing. Most of the errors don’t show up as problematic right away, and if the book is in print, it can be difficult or impossible for the writer to really recover. If you realize while writing Book 5 that you don’t need a subplot that you introduced in Book 2, and the first three volumes are already in print, you’re probably stuck. Even trying to revise Book 4 to downplay it may be difficult, depending where that book is in the production process.

Since very, very few writers are in a position to write four or more books entirely on spec (i.e., without a contract and without allowing any of them to be published until they’re all finished), this leaves prevention as the only option. Since this post is already a bit long, I’ll work on that on Wednesday.

 So my sister decided she wanted to see “The Hobbit” before she goes off on vacation with my Dad, and we rounded up the usual suspects and made arrangements for Friday, two days ago. After much discussion we all decided to meet in the middle (geographically speaking), which had the significant benefit of allowing us to go have Indian food at the good spot four blocks away from the movie theater.

 Lois and I had seen the movie the week before (and our reaction was to immediately come home and watch “The Fellowship of the Ring” on Lois’s TV). Setting aside technical questions about frame rates and the desirability (or not) of 3D filming, the discussion brought up a lot of interesting things about working with a series in both literary and visual formats, and the difficulties inherent in translating from one to the other.

 The first interesting point is that Tolkien wrote The Hobbit first, and at the time it was first published he did not know the significance the ring – and Gollum – would have in the later books. The movies were made in reverse order: The Lord of the Rings came first, and now we’re getting The Hobbit.

 This difference creates some interesting storytelling problems. The first is in tone. The Hobbit was written as a children’s book; it became an introduction to the epic trilogy that followed, but that was later. Moving from the tone of a children’s book to that of the adult fantasy is a little tricky, but only a little. It does, after all, follow the natural chronological flow, from child to adult, from children’s story to adult sequel.

 You don’t get the same effect, obviously, when you go the other way (from adult epic fantasy to children’s story), as the films do. The movie-makers chose not to try: the movie The Hobbit is filmed in much the same tone as The Lord of the Rings, and I can’t really see it working any other way.

 The question of tone blends into the question of continuity. The makers of the movies opted for continuity of tone and presentation movie-to-movie, rather than for consistency of book-to-movie tone and presentation.

 I’ve heard people grumble about this, but do bear in mind: the movie-makers had the choice. They had all four books right there in front of them before they ever started working on the first movie. Tolkien did not have that choice, because when he wrote The Hobbit he hadn’t yet made up all the things that came up later in The Lord of the Rings. Yes, he made some continuity changes to later editions of The Hobbit, but he could not have chosen to change the tone without doing a complete, massive rewrite of the book.

 It is, of course, possible that even if Tolkien had known the story was leading to The Lord of the Rings and the end of the age, he would still have chosen to write The Hobbit as a lighter children’s book. I take leave to doubt it, but authors have done stranger things. That choice, however, remains firmly in the realm of speculation, because Tolkien did not know. And I personally do not think that it would be right for a modern movie-maker to pretend that he is in the same position as Tolkien – that he’s making a children’s movie that those other books and movies don’t inform.

 Series continuity, whether in film or in print, is always a tricky business. Whether you write in chronological order, as Tolkien did with The Hobbit and then The Lord of the Rings, or whether you tell one story and then back up and write a prequel, as Tolkien did later with The Silmarillion, there will be people who encounter the story out of order. I read The Two Towers first, because it was the only fantasy on the airport book rack when my family was heading out on vacation (I knew perfectly well it was the middle book of a trilogy; I simply didn’t care). I then galloped through The Return of the King, followed up with The Fellowship of the Ring, and only then discovered that there was an earlier book called The Hobbit.

 Similarly, one very-much-not-a-fantasy-fan acquaintance heard that The Return of the King had been nominated for Best Picture Oscar, so he went blithely off to see it without having seen (or read) The Fellowship of the Ring or The Two Towers. Needless to say, he was deeply puzzled by the experience.

There is nothing whatever that a writer or a movie-maker can do to prevent this. You don’t have a choice in the matter. The only choices you have relate to how you tell the story: whether you try to make it accessible to people who may not have all the background, or whether you don’t.

As I said before, there are a lot of really good reasons for a multi-volume-novel type series to need a bit of review or reminder in the first couple of chapters (ideally in Chapter One, but you can’t always swing that).

None of those reasons apply to the stand-alone type series.

If what you’re writing is a collection of stand-alone novels, then each one ought to stand alone. That means that the writer doesn’t put in lots of extraneous-but-interesting information to get new readers up to speed, any more than the writer of a non-series stand-alone novel puts in a huge infodump recalling the protagonist’s life history to that point in his/her life. Everything that happened before Page 1 of the current novel is history, and the fact that some of your readers already know it doesn’t mean you have to treat it any differently from the background/backstory you made up prior to Book 1.

Yes, knowing the background and relationships that have developed over the past nine or ninety books ought to make the reader’s experience of the current book richer, but new readers don’t need to know all that in order to enjoy the current book. Book seventeen will be a different experience for people who come to it with sixteen books worth of backstory than it will for readers who haven’t read anything else in the series, but different does not mean bad or boring or unenjoyable. And it certainly does not mean incomprehensible – in fact, the amount of background a new reader requires in order to understand what’s going on is usually a lot less than the writer fears.

(This is, in fact, one of the potential advantages of writing a collection-type series. If the first two books of a five-volume novel are out of print when the last book hits the shelves for the first time, it’s a major problem. Very few readers want to read only the last half or the last third of a novel. If the first twelve books of a nineteen-book set of detective novels are OOP, it’s an annoyance for new readers who grab the latest one, love it, and want to go back and fill in, but it’s not the same kind of catastrophe.)

So, what goes into the book is however much background, backstory, or history needs to be there for that story, whether we’re talking about politics, the history of ancient China, the protagonist’s confused relationship with his/her childhood sweetheart, or the slowly growing friendship between the sidekick and the alien from Rigel VII. And rather than dumping it all in Ch. 1, you put the information in when the reader needs to know it – some in the first chapter, some in the third, some in the tenth…wherever it makes sense.

Note that “the amount of backstory that needs to be there” a) is nearly always less than the writer and faithful readers think; b) is not going to be the same for every novel in the series; c) is not related to where in the series a novel falls (Book 19, in which the heroine has been kidnapped by pirates and spends the entire novel dealing with them, may need very little of the background that’s been established in the 18 prior books, while Book 7, in which she’s dealing with a complicated plot to assassinate her best friend’s father-in-law [who’s also Chief Justice of the Interstellar Tribunal] may need to refer to nearly everything in the previous six books, one way or another); d) can vary if there’s a two- or three-book story arc mid-series, and e) often varies depending on stylistic and thematic considerations. In other words, like “it works,” how much backstory one needs is a judgment call.

Most writers have a fairly good handle on this when it comes to their characters’ history. You don’t see detective novels that start with a run-down of every murder the detective has solved in the past six books. What seems to trip people up most frequently are the character relationships. I’ve seen more than one great stand-alone series bog down around book six with what I call “check-in syndrome” – the writer spends more and more time at the front end of the book “checking in” on all the recurring characters the readers love, even if those characters have no particular part in the current story. Then the book either bloats up to twice the length it needs to be, or else the actual plot is crammed into the remaining few chapters, greatly to the detriment of the story.

What all this boils down to is that in a collection-type series, I’d recommend erring on the side of too little backstory rather than too much, unless you already know that you under-explain or unless you have solid stylistic or thematic reasons for running on and on about what’s already happened (for example, a garrulous first-person narrator…)

This one seems to give a lot of people fits, because there’s a lot of conflicting advice out there. And the reason there’s a lot of conflicting advice is, in part, because what you do with series backstory depends on what sort of series you are writing.

There are two basic kinds of series: 1) the sort that’s really a multi-volume novel, where plots and subplots carry over from one book to the next, and 2) the sort that’s a collection of different stories about the same set of people, with each book being a relatively independent adventure. A lot of the confusion about series backstory derives from mixing up the two types, because they have very different backstory requirements.

When you’re writing the first type of series, each book after the first one necessarily starts in medias res, because you’re really only telling one story that’s been broken up into parts. New readers coming in at book 2 or 3 will very likely be hopelessly confused if they don’t get some sort of summary of events to-date, and even fans of the series from way back may need a few reminders of what’s been going on, because it’s normal for the books to come out a year or more apart. Ideally, one wants this reminder as early in the book as possible, to minimize reader confusion and settle the story back into place.

The difficulty here is one of pacing. One doesn’t really want Chapter One of a book to be slow, no matter how many complex plots and subplots one needs to remind the reader of. One especially doesn’t want a too-detailed reminder of plots and subplots that will annoy people who come to the series after all the books are out and they can gulp them down one after another over a long weekend or vacation.

There are three basic methods for getting the necessary updates in anyway: one can treat the story as a more normal in medias res opening; one can set up the reminder in the previous book; or one can begin with a what-has-gone-before prologue, which readers can skip if they’ve recently read earlier books (or if they’re on their eighth re-read and know the plot by heart already).

The first method, writing as if the story is a stand-alone that just happens to start in the middle of the action, tends to work very well for a lot of stories. The writer doesn’t have to dump a huge mass of information into Chapter One; everything comes along just as the reader actually needs to know it, which for certain facts may not be until Chapter Ten. That means one can avoid a potential slow opening and get on with the story. Readers who dislike in medias res openings won’t care for it if they get hold of it first, but that’s why the publisher should put “Book 2 of…” in some clear and obvious spot on the cover.

The second method, setting up the summary in advance, requires careful planning, because the set-up usually goes at the end of the previous book – someone promises to write and explain, or demands a report, or talks about the upcoming conference, or mentions a new character who’ll need to be brought up to speed on what’s going on. The next book opens with writing the letter or the report or with the conference or the arrival of the new character, so the overall story flows along seamlessly.

When handled clumsily, this looks like the device it is – a compromise between working things in in bits and doing a prologue – and it’s not something you’d want to use book after book in a long-running series, because it gets really obvious if you do it more than once even if you’re clever. It can, however, work well if one is fairly far into a longer multi-volume novel with a complicated plot that even readers who have all the books ready to hand may be glad to see laid out clearly.

The third method, the what-has-gone-before prologue, is pretty self-explanatory; the main thing to remember here is not to get too attached to it, because people will undoubtedly skip it when they re-read it. It’s a practical solution if one has a convoluted plot, lots of subplots, and lots of characters, all of which the reader needs to be up to speed on before the opening scene.

The biggest problem most writers seem to have with all of these methods is a tendency to put in far more information than the reader actually needs to know in order to pick up where the story left off. (There are, as always, a rare few who put in too little information and end up making the reader’s potential confusion worse than ever, but as I said, they’re rare, and a good editor, beta reader, or crit group will call them on it.)

On the other hand, a multi-volume novel is still one story; it’s published in multiple books for reasons of production, or to give the reader a break (a million-plus words is somehow a lot more intimidating in a single volume than it is broken up into five or six more normal-looking books, even though nothing else about it changes). Deciding that one is not going to make any concessions to the publishing process (i.e., completely ignoring all of the above and putting in no what-has-gone-before stuff, nothing to remind a reader of what happened two books and three real-life years ago) is not a totally unreasonable position to take. One simply has to be very, very sure that one’s story is strong enough to hold up without extra artificial linkages between the parts, and willing to argue with or ignore all the editors, beta readers, etc. who tell you that you have to have something to fill in the reader on what has gone before.

As this is getting rather long, I’m going to stop here and do part 2, on handling the collection-of-stand-alones type series, on Sunday.

I have, however, had the opportunity to observe several writer friends as they produced various series and true multi-volume stories, and it seems to me that from a process standpoint, there are several types. Start with plotting: there are closed series and open series. A closed series is one that has a specific end in mind: after four or five or six or ten volumes, the protagonist(s) find the magical doohicky, recover the throne, save the universe, and it’s finally over. An open series is one that has the potential to keep going on and on, because there isn’t a background plot-arc that will ever be finished. The Enchanted Forest Chronicles in their current form are a closed series: the primary adventure is finished and the wizards are defeated for good. If I ever go back and write another of them, it’ll have to be about some completely new adventure, with a new antagonist (no, I don’t need suggestions, thanks… )  The Lyra books are an open series – they share a setting, but that’s really all. Lois Bujold’s Vorkosigan series is another example: Miles and his family are central characters, but each book is generally a separate adventure. There are a couple of story-arcs that cover two volumes, but none that string out for longer than that. Mystery series that follow a detective through various cases are another example.

Writing a book in an open series doesn’t differ much from writing a complete stand-alone, except that the writer has to pay a bit more attention to staying consistent with everything that’s been said in earlier books. (Unless the writer is Terry Pratchett, whose History Monks are possibly the most brilliant device ever thought up for justifying and explaining any and all background inconsistencies whatsoever.)

Writing a book in a closed series is a lot more complicated.  In most cases, and especially if the closed series is going to be longer than two or three books, the writer doesn’t want the current volume to read like a couple of chapters out of the middle of a stand-alone novel, yet progress has to be made toward the final series-ending goal or the series will stretch out to infinity without resolution and start losing readers after a while. This usually means that the book has to have a minor plot-arc or sub-adventure that can be finished up within one book, but that also moves the characters closer to their ultimate goal. And those two plot-threads have to be carefully balanced in the context of where the characters are in the whole series.

The structure of a long, closed series also needs careful attention (unless you’re the sort who does structure and plotting on a completely intuitive level and you can make this work over a multi-book project). The longer a closed series is, the more sub-arcs the story needs, to keep the reader interested and provide enough closure to avoid frustration (while not providing so much closure that they stop reading).

The other problem with a long closed series (one that runs for more than four to six novels) is that the writer frequently gets bored after four or five books if the end is still not really in sight. A bored writer is a dangerous thing. Quite often, this results in one of two things: either the writer starts branching out, doing side-stories and adventures for minor characters that have nothing to do with the central story-arc, or the writer sticks to the central story arc but stops paying attention. In the first case, all but the most dedicated readers eventually stop reading the series (since it isn’t getting any closer to a conclusion); in the second case, the quality of the books tends to drop, at which point, again, all but the most dedicated stop reading.

A good series, whether open or closed, provides a lot more opportunity for richness and complexity in both characters and background than a stand-alone novel. Plot…plot is another thing entirely. There aren’t really a whole lot of plots that need more than three or four books to tell, even in great detail.

Since this is getting a bit long, I’ll talk about specific tools and techniques for keeping tabs on a multi-volume story in another post

—

Pat said, “You’re going to kill me.”

That’s the way I remember my first encounter with THE MISLAID MAGICIAN. Pat Wrede and I were just finishing up with page proofs for THE GRAND TOUR – quite an undertaking, given that going over the changes involved conference calls with Pat, me, and Karrie Oswald, our Harcourt editor.

Pat and I were meeting for lunch. Pat’s warning accompanied a handful of pages. It was a letter to Kate from Cecy, and it became the opening of THE MISLAID MAGICIAN.

Pat rarely tells me things like “You’re going to kill me,” so I paid close attention as I read. It didn’t make me want to kill her. Far from it. It made me want to run home and write Kate’s reply to Cecy. So I did. And we were off.

As Pat has said in a previous post, the differences between our working styles made THE MISLAID MAGICIAN much more difficult to write than either of the other books. Although I thoroughly enjoyed tuning my internal receiver back to Radio Kate (and even more to Thomas, as we included letters from Thomas and James in this book), I had only a dim idea of what my half of the plot was going to be. I knew I wanted to include Drina. All else was mist.

My hat is off to Pat for cming so far out of her comfort zone to work with me so we could get the book done in a timely way.

We’d had a template for SORCERY AND CECELIA, and we had the map of Europe for THE GRAND TOUR. But we were completely on our own for THE MISLAID MAGICIAN. We had three advantages working for us, though.

The first advantage was returning to the epistolary style, writing letters back and forth between the characters. That was fun for us, and I think the reader can tell. The second advantage was the calendar. Pat constructed one when we were working on THE GRAND TOUR so that journeys would not conflict. For THE MISLAID MAGICIAN, she added moon phases and made it into a terrific tool for making sure the letters from four enthusiastic correspondents were posted at proper intervals and arrived no sooner than they ought to. The third and most vital advantage was our new editor, Kathy Dawson. She has a keen eye for detail and the instinct for the right word to a degree rare even among professional editors.

It’s been my privilege to work with Kathy Dawson again. She’s now at Dial Books. She edited MAGIC BELOW STAIRS, the middle-grade novel I have coming out from Dial in Summer 2010. The protagonist is Frederick, a servant boy in the household of Kate and Thomas. The book is set immediately after THE GRAND TOUR and the arrival of their first child plays a role in the plot. It is not epistolary, just a straight narrative, and it only concerns my characters, not Pat’s. But thanks to Pat’s gracious approval, it does take place in the same world as the Kate and Cecy books (my half of it anyway), and Thomas is an important character in the book. (Kate’s there too, I hasten to add.)

In the years since we finished THE MISLAID MAGICIAN, both Pat and I have been up to our eyebrows in independent projects. We have no plans to write another Kate and Cecy book. But I have to admit, while I was  writing this, I started wondering. Maybe someday I will hear Pat say,  “You’re going to kill me.” If I ever do, I will pay close attention.

Well, for starters, “getting a sequel” isn’t quite the right phrase. The first Kate and Cecy book, Sorcery and Cecelia, was put out by Ace/Berkley, who were not all that thrilled about the book at the time and most definitely did not want a sequel. So the book pretty much died, the first time around. Fast forward ten years…and the book has not only refused to die completely, but people are bidding over $100 on ebay for copies. Of a paperback.

So Caroline and I went to our agents and said, “What can we do with this?” And our agents said, “It’ll be easier to sell if there’s a sequel.” So we started writing the sequel, and sold it as a two-for with the possibility of a third.

The hard part about writing it was that The Grand Tour was so different from Sorcery and Cecelia. The structure of the sequel was pretty well locked in by the first book, both in terms of the alternating viewpoints and in terms of the setting – I suppose we could have skipped the story of the trip, but it would really have felt like cheating. And since the characters were all in the same place, we couldn’t have them write letters to each other, the way they had in the first book. We had to come up with an alternate conceit.

Also, Sorcery and Cecelia had the basic Regency Romance novel to use as a template for both plot and incidents. With The Grand Tour, we didn’t have a model to riff on. Most Romances don’t have sequels, and if they do, the “sequel” is really about other characters who were minor players in the original book, and the original couple(s) are treated as minor characters in the sequel. We wanted to keep writing about Kate and Cecy.

But the real problem with the sequel was that Sorcery and Cecelia was incredibly easy to write because we weren’t writing a book. We were playing a game. It ended up becoming a book, but we didn’t actually know that until the last few letters, and by then we were on the downhill run to the end. But we knew from the get-go that The Grand Tour was no game…and it made a difference. With Sorcery and Cecelia, we were perfectly happy to throw things in pretty much at random and see what happened, because the story didn’t have to come to any sort of conclusion (most of the other letter games that either of us have played just kind of ran out of steam after a while, usually in mid-plot). The Grand Tour needed to be going somewhere, which made us both a lot more self-conscious about what we were having happen.

The other big problem was that Caroline’s and my writing processes are almost pure opposites. She’s the sort of writer who can’t talk abut work-in-process; I’m the sort who likes to talk about it all the time to anyone who will listen. She doesn’t do plot outlines; I do (and then ignore them…but that, too, wasn’t as possible with this book because if I strayed too far from the sketchy plan we’d come up with, I’d throw her off). She generally underwrites in the first draft and then expands it in the second; I’m a rolling reviser, fixing previous scenes and chapters as I go forward, so by the time I’m finished with the “first draft” it usually doesn’t need more than a few tweaks. And so on.

Working out the plot and the “escalation” was fairly easy, compared to all that. Especially since, as I said, we didn’t have a template to riff on. About all we knew was that we did not want to do the thing you see in so many long-running TV shows, where they string out the romantic tension for years, then let the couple get together, then immediately try to regain the romantic tension by making the couple have a major misunderstanding or disagreement. Given that we were not doing that, we had to come up with something else to hang the plot on. Which we did, but it took a while.

So I’d say that in Sorcery and Cecelia, we paid more attention to the emotional/romantic plot as the main storyline, while the adventure part was more just the vehicle for making it happen. In The Grand Tour, we reversed that to some extent. Note that this is my analysis long after the fact, and not anything we consciously decided or verbalized at the time. It also doesn’t have a lot to do with the reader’s experience of the books. And Caroline has her own take on things, which obviously isn’t always the same as mine.

The first thing is that not all trilogies or series are the same. Some are accidental – the writer wrote a book that was supposed to be a stand-alone, and then everyone wanted more, more, more. Some are episodic – monster/murder/McGuffin-of-the-week (like the one Michelle is working on). Some are closed (or potentially closed) – there is one central story, and when that’s finished, the series is theoretically over (like “The Fugitive,” if anyone besides me still remembers that show). Some are completely open-ended.

Escalation is a problem for nearly any series that goes on long enough - the characters save the world (or whatever) and get stronger and better during the process, so saving the world again would be not only repetetive but too easy to be interesting. The writer has to come up with a new and bigger challenge – saving the universe, maybe. And that can get ridiculous pretty fast.

The escalation problem occurs most often and most obvioiusly when a writer does an accidental series. If the problem the characters faced in the original “standalone” book or trilogy was too large, it can be hard to find a new threat that doesn’t seem anticlimactic. If you know you’re doing a series, you can plan your challenges so that the characters aren’t saving the kingdom in Book 1 and saving the universe by Book 4.

The closed-ended series is subject to a slightly different kind of escalation problem. If you know that your characters are supposed to end the series by saving the world, you have to make sure that when they finally get to the climax of the series, it doesn’t turn out to be anti-climax. You have to either make sure that the characters don’t get too powerful over the course of their previous adventures, or else make sure that every time the characters get more powerful, the difficulty of saving the world (or whatever) increases, too. (Of course, in the series, one makes it look as if it was really going to be this difficult all along, only the characters didn’t know it yet.) And there’s always the danger of getting off-track and never actually completing whatever the original master-plot-arc was supposed to be, because one or more of the sub-arcs turns out to be more interesting (to the writer, at least).

Even the McGuffin-of-the-week type of series is vulnerable to the escalation problem. This is most obvious with TV series, because for just one full year of shows, the writers have to come up with about 30 plots, each with its own McGuffin…and beating the monster, catching the murderer, or finding the McGuffin can’t be too easy, or people get bored. So you start with the heroes beating random strange monsters, and by the end of a season or two, they’re fighting in the war between Darkness and Light, with the fate of the universe on the line, or they’re stopping Ragnarok at least once a season.

The open-ended series probably has the least trouble with escalation, because the author can and does switch to a new batch of characters if and when the first set gets too powerful. Worlds and universes are very large places; it isn’t that hard to find new characters who are far enough away that the overpowered folks from the first story aren’t going to find out about their problem in time to deal with it. And there’s always the next generation.

Avoiding escalation completely is really hard, if not impossible. The author is caught between a rock and a hard place:  if the characters have become stronger as a result of their adventures, the threats and obstacles they face must also become stronger, or the stories get boring pretty quickly. But if the characters have adventures and do not change, do not become stronger and wiser (or do so only at the proverbial snail’s pace)…readers also get frustrated after a while. About all you can do is pay attention and think ahead, and try to hit a pace of development for both your characters and your threats that is neither too fast nor too slow. (What?  Who said writing was going to be easy?)

Yes, I know, there are four Enchanted Forest books, and three Kate and Cecy books, and the Lyra series, and so on. But with all of those, I didn’t set out to write more than one book. I wrote a book, and then readers and editors asked for more.

This time, I signed on for three books right from the start. It seemed like a good idea at the time, and I don’t know that I can say I regret it, but it’s…different from the way I usually work. I have a notion of where the whole thing is going, rather like the notion I usually have of where and how I’m going to end a book, but it’s a lot more vague because it’s farther away.

And I’m having to be a lot more careful about things, because I know I’m not going to be able to go back and stick something into Book 1 if I happen to discover in Book 3 that I need it. Book 1 is in print; what’s there is there and I’m stuck with it. It’s kind of the extreme reverse of working without an outline (which I’ve done twice, and which always feels to me as if I’m doing a high-wire act without a net).

There’s another interesting thing, too:  whenever I’ve written books before, I’ve had no idea at all what they’re really about until they’re finished (and usually not until someone else points it out to me). The few times I’ve tried to identify a theme or a “what it’s really about” before I finished a manuscript, it’s a) been totally wrong and b) made it more difficult to finish. I do much better, I’ve found, when I can just say “Yes, this plot/character/background bit will work, but I’m sorry, that one just won’t. No, I don’t know why; it just won’t.”

This time…well, Book 1 is finished and out, and while I still can’t actually say “what it’s really about,” I have a much shorter list of possibilities than I usually would at this point in a manuscript. That worries me, because I have two left to write (see “more difficult to finish,” above). On the other hand, it made it quite obvious that a particular plot-point I was considering was completely bogus and didn’t belong in this book at all, because it didn’t fit with any of the possibilities. So perhaps there’s more of an up side than I thought.

In short, the whole writing process is, as usual, not following any of its previous patterns. I’ve been doing this for thirty years; I’m supposed to know how to do this by now, right? But apparently my backbrain has other ideas. That’s all right, though; as a rule, it is much smarter than I am.