The tyranny of the calendar

Note: I’m taking a few days off, so I’ll be republishing some of my favorite pieces from earlier in this blog’s run. This post originally appeared, in a slightly different form, on January 12, 2016.

When a novelist reaches a certain level of commercial success, the charge is inevitably leveled—as it still is against the likes of Stephen King, J.K. Rowling, and most recently George R.R. Martin—that he or she is no longer being edited. And it often seems like the evidence is right before our eyes. The books grow visibly longer, as they did most dramatically in the case of Harry Potter, or they take more installments to cover the same amount of ground, as with A Song of Ice and Fire. Familiar tics, like the folksy voice that King likes to assume, expand into full-blown affectations, and the novels themselves start to seem looser and shaggier. Something has clearly changed, and the underlying assumption is that the writers themselves are to blame: nobody likes being edited, and once their careers have advanced to the point where they carry sufficient financial clout with their publishers, they simply refuse to take any additional notes. As King himself said in an interview from the early eighties:

At this point, I think that if there were any change suggested to me that I didn’t want, all I would need to say would be, “No. I won’t do that.” And it would never be a question of their withdrawing my contract, would it? They’d just finally say, “Well, okay then, don’t do it that way. “Which means, in effect, that if I’m willing to be really intransigent, there’ll be no editing at all.

But the truth, as always, is a little more complicated. The quotation above comes from an interview with King first published in the second volume of the Dream Makers series by Charles Platt. It dates from an intriguing moment in King’s career, around the time of Christine, when he was already a force on bestseller lists but not the institution he later became. And he says of his editorial process during that period:

I like to write three drafts: a first, a second, and what I think of as the editorial draft, when I sit down and take an editor’s criticism and work it through in my own mind, and put the whole book through the typewriter again, and repolish the other stuff as well. But as the successes have mushroomed, it’s been tougher and tougher for me to get my editors to give me time to do that third draft. What I’m really afraid of now is that one of them will say, “I think this is great,” just because it fits their publication schedule. Every year, I’m on a faster and faster track…I am supposed to read the proofs [of Different Seasons] in five days. Now, what if we let a bunch of dumb errors go through? It isn’t a matter of creativity, or trying to do the best book possible, that’s governing things right now—it’s advertising. And that scares the hell out of me, because we’ll fuck up real good one of these days, and then people can say “Steve King writes for money,” and at that point they will be right.

This obviously reflects King’s own perspective on the matter, but it’s still a fascinating point, and it remains relevant when we flash forward more than thirty years to George R.R. Martin. In a blog post from 2009 titled “To My Detractors,” he recounts how he told his publishers that he wouldn’t be able to deliver the next book in the series on time, and he says of their response:

I thought they’d be sick about it…but I have to say, my editors and publishers are great, and they took it with surprising equanimity. (Maybe they knew it before I did.) They already had contingencies in place. They had made plans to speed up production. If I could deliver Winds of Winter by the end of the year, they told me, they could still get it out before the end of March.

Martin didn’t meet that deadline either, of course, and after describing his predicament in more detail, he concludes: “Best guess, based on our previous conversations, is that Bantam (and presumably my British publisher as well) can have the hardcover out within three months of delivery, if their schedules permit.” And although this line wasn’t much discussed in the fury of analysis that ensued, it may be the most astonishing tidbit in the entire post. Even if you just consider the physical challenge of printing a million hardcover copies, three months to take a novel from manuscript to bookstores is insane. With such a huge machine trembling to go into action, something’s got to give—and it’s probably going to be the editing.

What I’m suggesting, in other words, is that the perceived lack of editing in many big bestsellers isn’t due to authorial stubbornness or editorial laziness. Instead, it’s a structural consequence of fitting blockbuster books into a relentless publishing cycle. When you consider how the whole process is being squeezed on one side by the author’s pressure to finish and on the other side by the pressure to deliver the book to readers, it’s no surprise if certain crucial steps get truncated or eliminated along the way. And it makes sense that the first casualty would be editing. Authors often complain that no one really gets what they do, and that’s doubly true for editors. A process that is so opaque to outsiders is bound to fall by the wayside when there’s so much else to consider: you’ve got to drop something to keep on schedule, and it may as well be the editorial phase, which nobody understands anyway. (Which leads me to a crucial point that deserves a blog post of its own: this is also why tentpole movies these days seem to be consistently half an hour too long. There just isn’t the time to edit them properly.) If The Winds of Winter comes out three months after Martin delivers his “final” draft, there’s no way that it gets the edit it deserved: every other stage demands a fixed amount of time to complete, and it’s the edit that ends up paying the price. So when you worry that the books in your favorite series are getting longer and more self-indulgent, you don’t need to blame the editor or the author. You can blame the calendar.

I can dream, can’t I?

For years, I’ve been daydreaming about a piece of fan fiction that I’d love to write, although I doubt I’ll ever get a chance to do it. Let’s call it The Carousel. It’s a midquel to Inception, which means that it takes place during the events of the original movie—in this case, after Cobb has assembled his team for the mind heist, but before they’ve actually gone into Fischer’s head. (There’s nothing in the film itself to rule this out: it’s unclear how much time passes after Saito approaches them with the assignment.) Cobb is concerned about Ariadne’s lack of experience, so he proposes that they practice first with a quick, straightforward job. It’s a commission from a striking, mysterious woman in her fifties who wants them to enter her aging father’s dreams to discover the secrets of his past. She is, of course, Sally Draper from Mad Men. The rest of the story follows the team as they invade Don’s mind, burrowing into his memories of his life at Sterling Cooper and the women he loved and lost, and probing ever deeper toward the dark heart of the man who was once known as Dick Whitman. We’d see Arthur and Ariadne trying to blend in at the office holiday party, or maybe Eames going undercover in Korea. And when they emerge from Don’s brain at last, with or without the answers that Sally wants, they’ve all been subtly changed, and they’re ready to go after Fischer. If nothing else, it explains why they’re still wearing those suits.

Alas, I don’t think I’ll ever write this story, mostly because I know I can’t give it the energy and attention it deserves. After I got the idea for the crossover, I decided to put it off until Mad Men finished its run, which would allow me to draw on Don’s full backstory, but the longer I waited, the more obvious it became that I couldn’t justify the investment of time it required. For one thing, I’d want to write it up as a full novel, and to do it justice, I’d have to go back and watch all seven seasons of the series, looking for places in which I could insert Cobb’s team into the background, à la Back to the Future Part II. I’d also want to revisit Inception itself to see if there were any plot holes or contradictions I could explain in the process. In short, it would be a lot of work for a story that I’m not sure anybody else would read, or particularly want to see. But I seem to have incepted myself with it, because I can’t get it out of my head. As with most fanfic, there’s an element of wish fulfillment involved: it allows me to spend a little more time with characters I probably won’t see ever again. Mad Men ended so beautifully that any continuation—like the Sally Draper spinoff series that was pitched in all seriousness at AMC—would only undermine its legacy. And Inception is one of the few recent blockbusters that deliberately makes a sequel impossible, despite the occasional rumblings that we hear along those lines. It won’t happen. But this is why fanfic exists.

In the meantime, I’ll sometimes try to scratch that itch by reading a novel or short story and mentally casting all the characters with faces from Mad Men. It’s a habit that I picked up years ago, when I first read Arthur Hailey’s Hotel, and I’ve done it since with Airport and a few of John D. MacDonald’s novels. (I still think that Jon Hamm would make a perfect Travis McGee.) And the show maps onto George O. Smith’s stories about the space station Venus Equilateral almost too well. I’ll often do it when reading a story that is best approached as a period piece, thanks either to the author’s intentions or to the passage of time. Picturing Don, Joan, and the rest at least allows me to keep the clothes and hairstyles straight, which is a more significant factor than it might first appear: a book like John Updike’s Couples reads altogether differently when you realize that all of the women would have been dressed like Betty Draper. In other cases, it amounts to a hybrid form of fanfic, enabling the kind of dream casting that still makes me wish, say, for a miniseries version of The Corrections starring the cast of Arrested Development—which just makes me want to read that novel again with those actors in mind, just as I recently went back to Red Dragon while picturing Hugh Dancy as Will. It’s a harmless game, and it can bring out elements of a story that I might have overlooked, just as the casting of a particular movie star in a film can clarify a character in ways that a screenwriter can’t.

And this is just a variation on what happens inside all our heads when we read a novel. Only half of the work is done by the writer on the page; the other half occurs in the reader’s brain, which populates the novel with faces, settings, and images that the author might never have envisioned. What I see when I read a story is drastically different from what appears in your mind’s eye, and we have no way of comparing them directly. (That said, an adaptation can lock certain elements into place for many readers, so that their imaginations run more or less in parallel. Ten years ago, no two fans saw the characters from A Song of Ice and Fire in quite the same way, but thanks to Game of Thrones, I suspect that a lot of readers now just picture Peter Dinklage and Emilia Clarke, as if a wave function had collapsed into exactly one eigenstate.) The fact that fanfic bridges that gap instantaneously, so that we can immediately see all of our favorite characters, is a large part of its appeal—and the main reason why it’s a flawed school for writers who are still learning their craft. Creating believable characters from scratch is the single hardest aspect of writing, and fanfic allows you to skip that crucial step. Aspiring writers should be wary of it for the same reason that the playwright Willy Russell avoids listening to music or drinking wine while he works: “I think both those things seduce you into thinking that the feelings engendered by the wine or music are present in your work.” That’s true of fanfic, too, and it’s why I’ll probably never end up writing The Carousel. But I can dream, can’t I?

The tyranny of the calendar

When a novelist reaches a certain level of commercial success, the charge is inevitably leveled—as it still is against the likes of Stephen King, J.K. Rowling, and most recently George R.R. Martin—that he or she is no longer being edited. And it seems like the evidence is right before our eyes. The books grow visibly longer, as they did most dramatically in the case of Harry Potter, or they take more installments to cover the same amount of ground, as with A Song of Ice and Fire. Familiar tics, like the folksy voice that King likes to assume, expand into full-blown affectations, and the novels themselves start to seem looser and shaggier. Something has clearly changed, and the underlying assumption is that the writers themselves are to blame: nobody likes being edited, and once their careers have advanced to the point where they carry sufficient financial clout with their publishers, they simply refuse to take any additional notes. As King himself said in an interview from the early eighties: “At this point, I think that if there were any change suggested to me that I didn’t want, all I would need to say would be, ‘No. I won’t do that.’ And it would never be a question of their withdrawing my contract, would it? They’d just finally say, ‘Well, okay then, don’t do it that way.’ Which means, in effect, that if I’m willing to be really intransigent, there’ll be no editing at all.”

But the truth, as always, is a little more complicated. The quotation above comes from an interview with King first published in the second volume of the Dream Makers series by Charles Platt. It dates from an intriguing moment in King’s career, around the time of Christine, when he was already a force on bestseller lists but not the institution he later became. And he says of his editorial process during that period:

I like to write three drafts: a first, a second, and what I think of as the editorial draft, when I sit down and take an editor’s criticism and work it through in my own mind, and put the whole book through the typewriter again, and repolish the other stuff as well. But as the successes have mushroomed, it’s been tougher and tougher for me to get my editors to give me time to do that third draft. What I’m really afraid of now is that one of them will say, “I think this is great,” just because it fits their publication schedule. Every year, I’m on a faster and faster track…I am supposed to read the proofs [of Different Seasons] in five days. Now, what if we let a bunch of dumb errors go through? It isn’t a matter of creativity, or trying to do the best book possible, that’s governing things right now—it’s advertising. And that scares the hell out of me, because we’ll fuck up real good one of these days, and then people can say “Steve King writes for money,” and at that point they will be right.

This obviously reflects King’s own perspective on the matter, but it’s still a fascinating point, and it remains relevant when we flash forward more than thirty years to George R.R. Martin. In the blog post that I discussed here last week, he recounts how he told his publishers that he wouldn’t be able to deliver the next book in the series on time, and he says of their response:

I thought they’d be sick about it…but I have to say, my editors and publishers are great, and they took it with surprising equanimity. (Maybe they knew it before I did.) They already had contingencies in place. They had made plans to speed up production. If I could deliver Winds of Winter by the end of the year, they told me, they could still get it out before the end of March.

Martin didn’t meet that deadline either, of course, and after describing his predicament in more detail, he concludes: “Best guess, based on our previous conversations, is that Bantam (and presumably my British publisher as well) can have the hardcover out within three months of delivery, if their schedules permit.” And although this line wasn’t much discussed in the analysis that ensued, it may be the most astonishing tidbit in the entire post. Even if you just consider the physical challenge of printing a million or more hardcover copies, three months to take a novel from manuscript to bookstores is insane. With such a huge machine trembling to go into action, something’s got to give—and it’s probably going to be the editing.

What I’m suggesting, in other words, is that the perceived lack of editing in many big bestsellers isn’t due to authorial stubbornness or editorial laziness. Instead, it’s a structural consequence of fitting blockbuster books into a relentless publishing cycle. When you consider how the whole process is being squeezed on one side by the author’s pressure to finish and on the other side by the pressure to deliver the book to readers, it’s no surprise if certain crucial steps get truncated or eliminated along the way. And it makes sense that the first casualty would be editing. Authors often complain that no one really gets what they do, and that’s doubly true for editors. A process that is so opaque to outsiders is bound to fall by the wayside when there’s so much else to consider: you’ve got to drop something to keep on schedule, and it may as well be the editorial phase, which nobody understands anyway. (Which leads me to a crucial point that deserves a blog post of its own: this is also why tentpole movies these days seem to be consistently half an hour too long. There just isn’t the time to edit them properly.) If The Winds of Winter comes out three months after Martin delivers his “final” draft, there’s no way that it got the edit it deserved: every other stage demands a fixed amount of time to complete, and it’s the edit that ends up paying the price. So when you worry that the books in your favorite series are getting longer and more self-indulgent, you don’t need to blame the editor or the author. You can blame the calendar.

The song has no ending

Nearly seven years ago, when readers of A Song of Ice and Fire were anxiously awaiting the appearance of A Dance with Dragons, George R.R. Martin published a blog post titled “To My Detractors.” He noted “the rising tide of venom” that had arisen in response to the book’s lateness, and he wrote:

Some of you are angry about the miniatures, the swords, the resin busts, the games. You don’t want me “wasting time” on those, or talking about them here. Some of you are angry that I watch football during the fall. You don’t want me “wasting time” on the NFL, or talking about it here.

Some of you hate my other projects…Some of you don’t want me attending conventions, teaching workshops, touring and doing promo, or visiting places like Spain and Portugal (last year) or Finland (this year). More wasting time, when I should be home working on A Dance with Dragons.

After all, as some of you like to point out in your emails, I am sixty years old and fat, and you don’t want me to “pull a Robert Jordan” on you and deny you your book.

Martin obviously didn’t take such criticisms all that seriously. Last week, however, he published another post that was very different in tone. A Dance with Dragons had finally come out four years earlier, and fans had moved on to clamoring for the release of The Winds of Winter. Martin wrote: “You wanted an update. Here’s the update. You won’t like it.” He acknowledged that the book wasn’t close to being done, and he continued:

Unfortunately, the writing did not go as fast or as well as I would have liked. You can blame my travels or my blog posts or the distractions of other projects and the Cocteau and whatever, but maybe all that had an impact…you can blame my age, and maybe that had an impact too…but if truth be told, sometimes the writing goes well and sometimes it doesn’t, and that was true for me even when I was in my twenties.

This post was widely reported and analyzed, but few observers appear to have noted the extent to which it deliberately echoed its predecessor, almost point for point. Martin seemed to grant that the “distractions” invoked by his detractors might, in fact, have been partially responsible for the delay—and although this sounds like a concession to his critics, it feels more to me like an act of self-wounding from a writer who is already deeply depressed, to use his own words, by his own lack of progress.

And as much as I can understand it, it saddens me. Martin is a gardener, not an architect, and as an avowed architect myself, I can speak with some objectivity about the advantages, as well the disadvantages, of the gardener’s approach. What impatient fans sometimes fail to recognize is that the very elements that they love so much about the series arise from precisely the same place as the factors that have led to these delays. Its density of detail, its attention to character, its sense of taking even its author by surprise: all are inseparable from a creative process that is inherently unpredictable. In a blog post that most famously included the line “George R.R. Martin is not your bitch,” Neil Gaiman got close to the heart of the matter:

It seems to me that the biggest problem with series books is that either readers complain that the books used to be good but that somewhere in the effort to get out a book every year the quality has fallen off, or they complain that the books, although maintaining quality, aren’t coming out on time.

And the tradeoff between time and quality—which might strike regular readers of this blog as familiar—is especially true of a series like A Song of Ice and Fire. If you want to live with the richness and unpredictibility that the gardener provides, you have to be prepared to die by it as well. And if those double-edged qualities weren’t there, you wouldn’t have been drawn to these books in the first place.

But there’s also a very real sense in which the series’s own successes contained the seeds of its downfall. (This is certainly true of Game of Thrones itself, which I’ve stopped watching largely because of issues that were invisibly contained in its conception from the very beginning.) Martin’s approach to writing isn’t wrong, but it’s problematic when linked, like a conjoined twin, to a television series that has to release new seasons on a regular schedule. The idea of a novelist finishing a book series in parallel with its production in other media isn’t unprecedented: J.K. Rowling and Harry Potter did much the same. But the movies are less hungry for plot and more forgiving of delay, and big franchises, like the James Bond series, have weathered long interruptions in production without damaging the brand. A cable series can’t do that, and the pressure on Martin, which is clearly enormous, arises from a structural tension between the kinds of novels he writes and the implacable logic of television—which doesn’t even mention the pressure from his publishing house, which is a huge machine trembling to take action as soon as his manuscript is delivered. Martin, who spent years writing for television, knows this, but he still hoped he could make it work: “I never thought the series could possibly catch up with the books, but it has.” His disappointment in himself is painfully clear, and his sole consolation should be that what he was trying to do was probably impossible. Being unable to write to your satisfaction is the worst thing that can happen to any writer, regardless of the larger systems in which he plays a role, and we can only say to writer’s block what Arya is told to say to death: “Not today.”

Game of spoilers

Recently, I entered one of the most nervewracking phases of my pop culture life: I’m almost, but not quite, caught up on Game of Thrones. For various reasons, mostly because my diet of television was already overstuffed, I hadn’t gotten around to watching any of the series until last month, shortly after the fourth season premiered. When I did the math, I found that I could catch up quickly enough to watch the back half of the season as it aired if I ran through an episode a night, which is precisely how my wife and I have been spending the last four weeks. Fortunately, it’s the kind of show made for an epic binge, and there are extended stretches, especially in the second season, when you can’t wait to keep going. (Things slow down considerably in the third season, probably because of the decision to split one book across two ten-episode runs, but that’s a subject for another post.) And now that I’m only three episodes away from being able to watch the show in real time, I’m both relieved and a little anxious. Navigating the spoilers for the series so far has been an adventure in itself, and now that I’m so close to the goal, I feel a little like an explorer reaching the end of a series of booby traps—which, as Indiana Jones pointed out, is usually when the ground falls out from under your feet.

Game of Thrones is unique, of course, in that the spoilers theoretically never end. So far, the show has adapted two and a half books from a seven-book series, five installments of which have been published, which means that something like four more years of plot points are already in print. (If it isn’t clear by now, I’m coming to the series before encountering the books, although I expect that I’ll probably read them all before long, especially if I can figure out a good way to read A Feast for Crows and A Dance with Dragons simultaneously.) The series has shown a willingness to depart from the books in certain ways, but it’s fair to assume that it will closely follow the overall narrative as laid down in Martin’s novels. As a result, viewers of Game of Thrones have a different relationship to spoilers than with any other show in history. On the one hand, if you’re just a day or two behind, there’s an enormous risk of spoilers online and in social media: even if you manage to stay off Twitter while the current episode airs, there’s always the danger of running into a gif, an offhand reference, or a think piece in the New York Times. On the other hand, beyond that immediate risk, we have what you might call the Experts and Newbies divide, and by the time we’re all on the same page, we’ll be well into the next presidential administration. And even knowing roughly how many years the series is going to run counts as something of a spoiler in itself: no matter how far along the narrative seems, we know that there’s much, much more to come.

Yet there are also ways, perversely, in which having certain developments spoiled has enhanced my enjoyment of the show. I nearly made it to “The Lion and the Rose” with one big twist still unknown to me, only to have it spoiled the day before by a gratuitous joke in an unrelated article on The A.V. Club. Not surprisingly, I was pretty mad about this—I was so, so close—but I found that it didn’t diminish the episode’s suspense. In some respects, it increased it: Hitchcock famously describes the difference between suspense and surprise, coming down squarely in favor of the latter, and much of my tension during the episode came from my knowledge of what was coming, much as Vertigo works better when its central revelation arrives at the beginning of the third act, rather than at the end. Did it make the episode better? Not necessarily, and I’m sure the creators would have preferred, as I did, that I’d gone in cold. Still, it allowed me to appreciate how cleverly all of the elements of the story were being laid into place, which otherwise would have had to wait until a second viewing, and it didn’t lessen the impact of a virtuoso set piece which ranks among the best things the show has ever done.

And there’s a sense in which our prior knowledge of the show—whether through accidental spoilers or having read the books—adds another dimension to the viewer’s engagement. The best parallel to this is Hannibal, a show based on source material that I do know by heart, and which pushes against its inspiration in ways that have become delightful and surprising. Even the most casual viewer probably knows more or less where Hannibal Lecter’s story is going, and Bryan Fuller’s stated intention to carry the series through Red Dragon, The Silence of the Lambs, and the climactic Hannibal novel itself tips us off to the arc of the stories for years to come. So far, however, this hasn’t been a weakness but a strength: Fuller has shown a readiness to depart from canon in startling ways, as well as to incorporate new arrangements of the puzzle pieces he has, and for a Thomas Harris fan, that’s part of the fun, whether it comes from a line of repurposed dialogue or a new interpretation of a character we thought we knew. We may have a good idea of the what, but the how remains unknown. And it’s in the how, or in those aspects of style and inspiration that can’t be spoiled in advance, that great television thrives.

“She couldn’t believe it was over…”

Note: This post is the twenty-second installment in my author’s commentary for City of Exiles, covering Chapter 21. You can read the earlier installments here.)

I’ve spoken frequently on this blog about my fascination with television, and especially with the challenges involved in telling an extended, episodic narrative in a medium that offers no clearly defined endpoint. Trying to plan too far in advance is a fool’s game: you never know if you’re going to get a single pilot, ten episodes, or a decade’s worth of stories, so you simultaneously need to act as if you might be canceled tomorrow and if you’ll get six seasons and a movie. Inevitably, the situation encourages shows to burn through ideas as quickly as possible. Since you don’t know how long you’re going to be on the air, it doesn’t make sense to hold any of the good stuff in reserve—which is why so many series seem to work themselves into exhaustion around the fifth season or so. At the other extreme, planning too far in advance can rob a show of its surprise and spontaneity, as I argued last year in my Salon piece on The X-Files and House of Cards. Even a show like Breaking Bad, which was able to set its own timetable for its final run of episodes, allowed chance and uncertainty to creep into the process: when Vince Gilligan and his writing staff showed us that machine gun in the trunk of Walt’s car at the beginning of the final season, even they weren’t sure how it was going to be used in the end.

Since a book is written and conceived in its entirety before its initial publication, it’s harder for a novelist to set challenges like this, although there are exceptions. Writing in a serial format, as Tom Wolfe did with the original draft of The Bonfire of the Vanities and Stephen King did with The Green Mile, can create something of the same effect, and it’s an approach I’ve often been tempted to try. It’s also a factor for authors of series fiction. Plotting out one novel is hard enough in itself, much less trying to figure out a narrative arc that spans several volumes, and I suspect that most writers who stretch a single story across multiple books aren’t quite sure how the final result will look. George R.R. Martin’s evolving sense of the scope of A Song of Ice and Fire has been thoroughly documented—he originally conceived it as a trilogy, only to see it balloon to a projected seven installments—and although he claims to have a general sense of how the story will end, I expect that even he will be surprised by many of his own developments. Part of this has to do with sustaining the writer’s interest over a project that can consume many years of his or her life, but even more of it has to do with the impossibility of holding that much information in one’s head at any one time.

As a result, a plot development in the middle of a series can sometimes resemble a shot in the dark, a best guess as to which avenues of exploration will turn out to be dramatically profitable. By the time I started work on City of Exiles, I knew that I wanted it to be the middle volume of a trilogy, which meant that every choice would affect not just the plot I was writing, but a hypothetical third book whose outlines I could see only dimly. (I’d like to believe that the three books in the series feel like one unified story, but I had no idea what Eternal Empire would be about, or even who the protagonist would be, until I’d submitted a draft of the second installment.) Of all the judgment calls I made, the one that had the greatest impact on what followed was the decision to have Ilya captured by the police at the exact midpoint of the series. At the time, as I mentioned last week, it was a way of forestalling writer’s block: I didn’t want to write the same novel all over again, and the change was radical enough to get me excited about where the story would go. But it also imposed enormous limitations, since I was effectively confining my most dynamic character for an extended period of time, and it meant that I was suddenly writing a prison novel, which wasn’t exactly what I’d had in mind when I started.

When I wrote the scene of Ilya’s capture, which occurs in Chapter 20, I knew that all of these elements would present issues down the line, but I tried to take a page from the episodic narratives I admired and focus solely on the moment. I think the result is an exciting scene, and hopefully an unexpected one, all the more so because I didn’t know what would happen next. Looking back, it’s hard to imagine this series unfolding in any other way, and it certainly opened up a rich vein of new ideas. (For what it’s worth, I don’t think I would have been brave enough to write this scene at all if I didn’t know that I had ample background material available on the British prison system, in the form of memoirs, journalism, and other works of nonfiction, most of which will make its appearance in Part II.) By writing both Ilya and myself into a corner—and one I wouldn’t be able to get out for another three hundred pages, spread over two different books—I gave the series a jolt of energy that it badly needed. Whether the result was better or worse than it would have been if I’d gone in a different direction is something I’ll never know. All I can say is that it made me a lot more curious about the outcome, and that when I put Ilya in handcuffs, I had no idea how, if ever, he’d get out of them again…

George R.R. Martin on world-building

[George R.R. Martin, author of A Song of Ice and Fire] enjoys being surprised by his own work. He thinks of himself as a “gardener”—he has a rough idea of where he’s going but improvises along the way. He sometimes fleshes out only as much of his imaginary world as he needs to make a workable setting for the story. Tolkien was what Martin calls an “architect.” Tolkien created entire languages, mythologies, and histories for Middle-earth long before he wrote the novels set there. Martin told me that many of his fans assume that he is as meticulous a world-builder as Tolkien was. “They write to say, ‘I’m fascinated by the languages. I would like to do a study of High Valyrian'”—an ancient tongue. “‘Could you send me a glossary and a dictionary and the syntax?’ I have to write back and say, ‘I’ve invented seven words of High Valyrian.'”

—Laura Miller, in The New Yorker