“May I please speak with Maddy Shaw?”

Note: This post is the second installment in my author’s commentary for Eternal Empire, covering Chapter 1. You can read the previous installments here.

In the classic mystery companion Murder Ink, the novelist Peter Dickinson draws a helpful distinction between two kinds of series characters. The first, or what he calls the “deliberate” creation, is conceived explicitly to carry more than one novel, and the author designs him with that role in mind:

They will have to have some kind of trademark. The hero will be different. Thus traits of difference are accumulated, selected not for the way they grow out of the character but solely because nobody else has yet thought of them…So the deliberate hero is jumbled into being…Let’s say he has a club-foot and rides an enormous bike…and carries a…a swordstick? No, too ordinary; what about a blow-pipe? And he knows the Bible by heart, huh?

The other sort of character, the “accidental” kind, is created for the needs of a particular book, and he or she often has traits that you’d never include if you were thinking in terms of a series. “The book itself demands a detective,” Dickinson writes, “and he grows into being, quite slowly, finding his shape and nature from the needs of the book and the author’s own needs. He may turn out a very odd creature, but all his oddnesses are expressions of what he is like inside.”

It’s the second kind of character who often ends up being the most interesting to read about, but also the most problematic to write. When the time comes to plug an accidental protagonist into a new story, the author often finds that he’s inherited a set of characteristics or a backstory that would never have been there if he were starting from scratch. Yet it’s often those instrumental, pragmatic touches, which arose in the first place because a different story demanded it, that push the plot in productive directions. It becomes another creative constraint, which is always a good thing, and it generates ideas precisely because it limits the writer’s options. In the case of my own novels, I’ve noted before that the decision to make Rachel Wolfe a Mormon, which ended up being central to the last two books of the series, was a random inspiration designed to fill out her character a bit: it was a late interpolation into The Icon Thief, and according to my notes, I seem to have briefly considered making her South Asian instead—an idea that dimly prefigures Maya Asthana, who evolved in crazy ways of her own. And I don’t think that Wolfe would have turned out half as interesting if I’d coolly constructed her with an eye to subsequent novels.

This is even more true of Maddy Blume, the protagonist of The Icon Thief, who was unexpectedly recruited for a similarly central role in Eternal Empire. If anything, her case is even more complicated: as the lead character in the first novel, she arrives with a lot of history and emotional baggage, all of which had to be acknowledged in the last installment without overwhelming readers who were encountering her for the first time. With Wolfe, I could take the few scraps of information I had and spin them into something largely new, while Maddy existed in a particular form that had been explored for hundreds of pages. The challenge of Eternal Empire, especially in its opening sections, consisted of reintroducing her, grounding her in a new situation, and then yanking her out of it. And it was especially difficult because Maddy isn’t the kind of person you’d expect to see in more than one suspense novel. A law enforcement officer like Wolfe will take on many cases in the course of her career; even a criminal like Ilya lives the kind of life that consistently courts danger. Maddy is a fairly ordinary young woman who, by the end of the first book, wants nothing more than to keep her head down, and to get her back into the story, her past had to come back to haunt her.

As a result, the opening chapter of Eternal Empire has to do about five different things at once, and on reading it over again, I think it pulls it all off pretty well. For reasons of plot, I needed Maddy in London, and the idea that she’d be working in the art world under a fake name seemed like a fairly logical step. (The name she takes, Maddy Shaw, is a reference to the pseudonym that T.E. Lawrence assumed when he tried to keep a low profile after the war.) The events described in the prologue provided a convenient way of blowing her cover, leaving her at a low point personally, professionally, and financially, and therefore receptive to the outrageous offer that Powell is about to make. And much of the subsequent plot, especially in the first half, was designed to make her entry point into the story—working as an art advisor to a Russian oligarch, while secretly gathering information about his financial activities—marginally plausible. It all required me to think a lot harder, and make things more difficult for myself, than if I’d started with a new character, and it influenced the overall shape of the novel in countless ways. Maddy, like the novel itself, is a very odd creature. And both she and it are more interesting, at least to my eyes, than they would have been otherwise…

“Why didn’t you put a red dress on her?”

I once ran into a very funny situation when Selznick, with great pride, showed me the big scene from Gone With the Wind. It was the scene with all the soldiers lying in the station yard and there was this high pull-back showing Vivien Leigh looking for her man. She was wearing a pale violet dress and you could hardly see her. I said, “David, why didn’t you put a red dress on her? When the camera finally reached the high point, all you would have seen was this little red dot.” That shook him. He had never thought of it. Ridiculous not to think of a thing like that. He missed the whole point of the scene entirely. And wouldn’t that have made some retake?

—Alfred Hitchcock, in Murder Ink

The challenges of series fiction

It would be nice if writing a series of novels came down to publishing a bunch of books with vaguely similar covers, but unfortunately, it isn’t that easy. Series fiction, especially in suspense, is inherently problematic because it violates an important aspect of what we know about the world. It’s true that life doesn’t lend itself to being confined easily within four hundred pages, and any story can be extended indefinitely in either direction—life rarely affords us the luxury of tidy endings. (As Sydney Pollack points out in Eyes Wide Shut: “Life goes on. It always does. Until it doesn’t.”) But series fiction pushes this sense of continuity in artificial directions, by assuming that not only does life go on, but it rhymes. In most mystery series, the protagonist doesn’t change appreciably from book to book, is confronted with the same kind of case in each installment, and usually ends up more or less where he started. Events in one book rarely have any impact on the next. And while this sort of structure is acceptable on television—although it took me a while to accept the lack of narrative memory on shows like The X-Files—it often rings false in fiction, and can even become enervating for the author himself. Conan Doyle famously killed off his own creation, and one occasionally senses a touch of exhaustion in such otherwise excellent writers like Daniel Silva, whose publishers have gently nudged him back toward his trademark character when he might have preferred to move on.

When it came time to write the sequel to The Icon Thief, I wanted to avoid this trap as best as I could. I was determined, for instance, that City of Exiles change the stakes of the series in tangible ways, and when I realized, early in the planning process, that I was simply repeating the same pattern as before—of Ilya staying just barely ahead of his pursuers—I decided to move forcefully in the opposite direction, with consequences that readers of the novel will discover for themselves. In some ways, I was aided by the fact that I’d never intended my debut novel to be the first installment in a series. In an excellent essay published in the classic mystery companion Murder Ink, the novelist Peter Dickinson draws a useful distinction between novels that were conceived with a series character in mind and those that stumbled into it by accident. A deliberate series hero, he observes, often starts out with a list of quirky character markers designed mostly to set him apart from similar protagonists (“Let’s say he has a club-foot and rides an enormous bike…”), while the accidental hero evolves in a more organic way, based on the needs of the first novel in which he appears:

These are the detectives who come into existence because the author wants to write a particular book. The book itself demands a detective, and he grows into being, quite slowly, finding his shape and nature from the needs of the book and the author’s own needs.

This is essentially what happened with Ilya, as well as with my beloved Rachel Wolfe. I didn’t intend to bring them back, and when I did, I found that I was stuck with the attributes I’d invented for them in the first novel—which ended up being a source of fruitful ideas and constraints. The single greatest trick I’ve learned from writing a series involves finding where the real essence of a novel, or a character, lies, and not confusing this with more superficial qualities. Ilya was conceived as a resourceful thief and assassin, but in the second novel, he doesn’t steal much of anything and kills only out of necessity. As a result, I was forced to dig deeper, and by following some hints from the first novel—his bookishness, his fascination with Jewish mysticism, and his discovery that everything he believed about his sense of honor was a lie—I was able to see him more clearly than before, as a man trying to come to terms with the division between the two halves of his personality. Similarly, if my only goal had been to write a novel with a story resembling that of the first, I would have centered it on another enigmatic work of art, which was the last thing I wanted to do. Instead, I looked at the novel slantwise, and saw that it was really about the problem of interpretation, and the stories we tell about ourselves and the world around us. This led me to concentrate on historical mysteries instead, and ultimately led me into the Dyatlov Pass.

Series fiction also turned out to be a testing ground for my ideas about backstory. Usually, I deal with my aversion to backstory by refusing to create it in the first place, to the point where important details of my characters’ backgrounds are left deliberately vague, even to me. For my second book, not only did this backstory already exist, but there was an entire novel’s worth of it, so I had to decide early on how much of this previous material to include. My solution, which probably won’t come as a surprise to regular readers, was to refer to the earlier book as little as possible, which ended up being both a philosophical and a pragmatic decision. The second book in a series occupies a peculiar position: it needs to reward those who have read the first novel, but also be accessible and interesting to those encountering these characters for the first time. Finding the right balance between telling a completely self-contained story and honoring the complicated history of the first novel was a real challenge, and it taught me a lot about what information is essential and what can be safely dropped—a lesson that I’ve put into practice for the third and final novel in the series, Eternal Empire, and hope to draw upon for any books I write in the future. The result, I’d like to think, is a pair of novels that read perfectly well on their own, but gain additional richness and resonance when taken as part of a larger whole. Tomorrow, on the day City of Exiles finally comes out, I’ll try to pull all of this together, and explain what I learned from writing my second novel.