Interview with Erika Dreifus

Erika Dreifus—fiction writer, reviewer, blogger, and self-described “resource maven”—recently published a short story collection called Quiet Americans (Last Light Studio, paperback, $13.95) that is profoundly historical in nature. Borrowing in part from her own family’s history, the book demonstrates the long term effects of the holocaust, not only on those who lived through it but on those later generations who find themselves in the United States only because in the 1930s an ancestor escaped Nazi Germany. For a fuller description of Quiet Americans, see my review of it on this blog.

Given that Erika is an experienced writer of historical fiction, and someone who has even taught classes on the subject, I wanted to interview her and capture her thoughts on some sticky questions related to this popular—but sometimes contentious—genre.

First, a simple, or maybe not so simple, question. How do you define historical fiction?

It's not so simple!

The Historical Novel Society offers a definition that I have found useful in launching these discussions (in a past life, I taught writing workshops for historical fiction writers):

"To be deemed historical (in our sense), a novel must have been written at least fifty years after the events described, or have been written by someone who was not alive at the time of those events (who therefore approaches them only by research)."

The organization goes on to say:

"We also consider the following styles of novel to be historical fiction for our purposes: alternate histories (e.g. Robert Harris' Fatherland), pseudo-histories (e.g. Umberto Eco's Island of the Day Before), time-slip novels (e.g. Barbara Erskine's Lady of Hay), historical fantasies (e.g. Bernard Cornwell's King Arthur trilogy) and multiple-time novels (e.g. Michael Cunningham's The Hours)."

Thank you. These are very useful distinctions, and, as the Society points out, all are legitimate, if varied, examples of historical fiction. How would you apply the definitions to your own book?

My work, as reflected in my new collection, Quiet Americans, is definitely more in keeping with the first, more "realist" part of the HNS definition. Three of the seven stories take place before my lifetime; a fourth is set during my very early lifetime (and was therefore depended entirely on research for the historical setting).

But while we’re on this topic, let me go a bit further on the issue of definition: I've long been intrigued by the way in which certain fictions written close to the time of the events they describe become "historical fiction" for the readers they reach many years later. For their authors, they may most accurately be considered "contemporary" or "political" fictions, but for the reader generations later, they exude historicity. For example, the first section of Irène Némirovsky's Suite Française provides what was a contemporary account of Paris in 1940, but today's readers may perceive it as "historical" fiction. When does the contemporary become historical? Some of the later stories in my book incorporate events that were "contemporary" when I was writing about them in 2004 or 2006. Are those stories already "historical" for the reader? Will they be more "historical" for a reader fifty years from now? These are tantalizing questions.

Yes, I think this is important. It gets at the idea of a fiction’s historicity stemming from the uniquely interesting/important time period in which it is set, be that far from the writer’s own time or close to it. I agree completely with your example of Suite Française. It’s impossible not to feel that a big part of the book’s intrigue is in how it portrays that crucial period in French and world history. I think too of Doctorow’s The Book of Daniel. No one would have called that book historical fiction when it first came out, but how can we not read it now with one eye on what it shows about the history of leftist politics in America?

Back to Quiet Americans. The book spans a wide stretch of time, from the early 20th century almost up to the present. Yet there is an obvious thematic connection between the stories. At what point did you know that you were composing a linked collection as opposed to separate stories? Did you ever consider turning the material into a novel?

Well, the stories certainly are linked thematically, and a few of them are linked further by characters and families who reappear from story to story, but some might argue against characterizing the book as a "linked collection," simply because not all of the stories involve the same characters/families. Which is all a prelude to saying that I'm not certain that I ever knew I was composing a linked collection, and I never seriously considered turning the material into a novel (perhaps because I had already written one unpublished, Holocaust/World War II-focused novel manuscript).

It seems important to note that the "oldest" of the stories in this book dates from a fall 2001 draft; three of the seven originated as submissions for MFA program deadlines. One of my program's strengths was its emphasis on generating new work: We were required to submit 8-25 pages of fiction twice during each semiannual residency and four times each semester. Revisions were acceptable, but even so, I wrote a lot of new stories in those years. Which means that I wrote a lot of stories that do not appear in this book. And shaping a collection was a process that took many years. At some point, I became certain that I had sufficient stories that cohered in some way to compose a collection—it just took me a long time to develop the particular content and sequence of Quiet Americans.

Wow, that’s a lot of composing over a very compressed time. The way in which it paid off for you is a good lesson for any writer. In several important ways, Quiet Americans draws directly from your own family's history of emigration to this country. Does exploring and utilizing one's own family's history affect the nature of writing historical fiction? Does it become harder or easier to insert oneself into past periods? Are there any extra burdens that you carry?

What great questions. I'm not sure that I can answer them right now. I'll want to think about them for quite awhile.

Overall, I've considered it an immense privilege to write these stories. The one pattern I'm noticing now that the book is out—I wouldn't call it a burden—is that I'm being asked by readers-who-are-family-members what, exactly, I've made up and what is "real" when it comes to the characters who most closely resemble my grandparents.

Yes, how often do we get asked that by our relatives, no matter what kind of fiction we’re writing? And the maddening thing is that they’ll never believe your explanations, because no one can who hasn’t immersed herself in the creative process. How much research did you carry out before starting the stories in Quiet Americans? How about other historical fictions you've written? How much of that research finds its way into the stories? And does your background as a historian give you an advantage?

In general, I've found that most of my historical fiction springs from some sort of osmosis, whether from having listened to and thought about various pieces of family history or having stumbled on a document or historical tidbit quite unintentionally. As I write, the research becomes increasingly important, but it's not usually the spark. And, like pretty much any other historical-fiction writer, I've uncovered plenty of material that ultimately doesn't make its way into the work.

I'd say that my background as a historian helps in several ways. For starters, I have a love for research and I'm not afraid to go looking for what I need. I'd also like to think that my training helps me approach and evaluate sources knowledgeably.

When you write a historical fiction are there any aspects of the past period that you feel are especially important to reproduce? For instance, settings or costume or diction? Are there any aspects you pay less attention to?

Another great set of questions. I do want everything to be plausible, but I probably pay less attention to settings, costume, and diction than others do. Some examples of historical details I've attended to quite carefully are the legal constraints that faced Jewish doctors in Nazi Germany (and then refugee doctors in the United States) in "For Services Rendered," the chronology of the Munich Olympics and the murder of Israeli athletes in 1972 for "Homecomings," and, in my unpublished novel, the medical protocols for managing the care of infants born prematurely around 1940.

I heard Ron Hansen say at one AWP session on historical fiction that when a writer is portraying an actual figure from history, he should not “knowingly depart from fact.” Do you accept this proscription?

I wish that I'd been there to hear Hansen say that, hear what prompted him to say that, and hear any responses. The use of "real people" in fiction is such a complicated issue. It always came up in my workshops on historical fiction, and some of the discussion always took place around an assigned reading of an edited transcript of a 1968 panel discussion that had taken place among Ralph Ellison, William Styron, Robert Penn Warren, and C. Vann Woodward.

And I always liked to quote Ellison, who argued that because "the work of fiction comes alive through a collaboration between the reader and writer," the dilemmas become more acute when fictionalizing individuals from more recent history. In contrast to an historical figure in a Shakespeare play, for instance, he suggested, "[I]f I were to write a fiction based upon a great hero, a military man, whose name is Robert E. Lee, I'd damn well be very careful about what I fed my reader, in order for him to recreate in his imagination and through his sense of history what that gentleman was. Because Lee is no longer simply an historical figure. He is a figure who lives within us. He is a figure which shapes ideal of conduct and of forebearance and of skill, military and so on. This is inside, and not something that writers can merely be arbitrary about. The freedom of the fiction writer, the novelist, is one of the great freedoms possible for the individual to exercise. But it is not absolute. Thus, one, without hedging his bets, has to be aware that he does operate within an area dense with prior assumptions."

And I also liked to quote Styron (who, it should be noted, was quite the center of attention at the time for his Confessions of Nat Turner), who presented this view: "[A] novelist dealing with history has to be able to say that such and such a fact is totally irrelevant, and to Hell with the person who insists that it has any real, utmost relevance. It's not to say that, in any bland or even dishonest way, a novelist is free to go about his task of rendering history with a complete shrugging off of the facts....It is simply that certain facts which history presents us with are, on the one hand, either unimportant, or else they can be dispensed with out of hand, because to yield to them would be to yield or to compromise the novelist's own aesthetic honesty. Certain things won't fit into a novel, won't go in simply because the story won't tell itself if such a fact is there....The primary thing is the free use and the bold use of the liberating imagination which, dispensing with useless fact, will clear the cobwebs away and will show how it really was."

It really is complicated. It really does depend. Did I depart knowingly from fact in "For Services Rendered"? I'm not sure. According to the facts, as I knew and researched them, "For Services Rendered" is entirely plausible. Is it factual? Most unlikely. On the other hand, the only words I put in Golda Meir's mouth in "Homecomings" are words that I am sure, from research, that she really said.

Thanks for the great quotes, and the insights. I can see how both Ellison and Styron, from their different perspectives, were responding to Nat Turner. I tend to lean toward Styron's view. Finally, of course, an adherence to fact is a very personal decision by the author, as your answer suggests. I don't like or want to just disregard facts, as Styron allows, but neither do I want to feel chained by them. Writing a novel is writing a novel, not writing a biography. There has to be a difference. As long as the author is open about what he's doing, and doesn't pretend to be strictly factual. Styron never did. Anyway, what you said about "For Services Rendered" gets to the heart of the matter for me. Even if a writer doesn't knowingly depart from fact, what she writes can still be extremely speculative and even implausible, fully a creation of her imagination. For the most part, that describes my Van Gogh novel, although I did depart from fact on occasion.

A different question. I know that some writers of historical fiction operate from the premise—or feel that have to—that while modes of external behavior (how people dress, how they talk, how they vote can change drastically over time) humanity remains essentially the same on the inside. Is that a premise you accept?

For the most part, yes, I have accepted that premise. Back in 2001, I read a wonderful essay by Geraldine Brooks on this topic, and I embraced what Brooks had to say wholeheartedly. But now, watching so many changes in the way we live and interact with each other—yes, technology has a lot to do with this—I have a few more doubts.

And before I sign off, please let me thank you, John, for inviting me to answer these questions, and for maintaining such a wonderful resource here at Creating Van Gogh for those of us who write historical fiction.

Thanks, Erika. Your comments were really useful. Good luck with your book!

Historical fiction's double identity

As my Van Gogh novel developed over the past four years, so did my attention to historical fiction. And what strikes me as one of the great curiosities about form is that it both can and can't be assigned the "genre" label. With the possible exception of science fiction, I'm not sure there's a fictional genre out there that leads such a double creative life, that has such a schizophrenic reception among readers and writers. On one hand, for decades historical fiction has been the locus for writers--many of them, let me say, perfectly hard working people--who aren't really intent on or concerned about creating books that can be lauded as "literary" so much as books they contribute to already existing and familiar genres. Many historical novels, for instance, were and still are little more than gussied up mass market romances or adventure books. There is too an abiding and popular genre of historical mysteries. And, of course, plenty of authors have written historical novels for children and young adults (some of them fine, and occasionally classic, books). While authors who write such books often do carry out quite extensive and valuable period research--research that does find its way into their novels--the end products are the type that cause historical fiction to get tossed into that long list of typically sneered at genre fictions. You know the list. You've seen it in every discussion of the literary marketplace, and in every journal's description of what it does or (more likely) doesn't want: romance, western, horror, suspense, children's, fantasy, sci-fi, sports, mystery, crime, etc. Now it goes beyond the purposes of this post to debate whether we ought to sneer at such genres at all--I know a lot of awfully smart people who say we shouldn't--but it's safe to say that the genres do earn sneers, even in this era of the ubiquitous, bestselling vampire novel that delights and consumes (no pun intended) so many people, even when so many stylish young writers (I see them in my classes) are absorbed by and committed to writing fantasy.

But here's the thing about historical fiction. Even while it lives out a full, happy, abiding life among the genres, at the same time it is embraced--increasingly so--by many "literary" writers. (I don't like the term, but for convenience sake I have to use it.) Don't get me wrong. There have always been acclaimed historical novels. But I feel an especially keen interest in the form now, even among younger authors, which certainly would not have been true in the past. Every year at conferences like AWP sections, sometimes multiple sections, on historical fiction are featured and are well attended--and not by anyone wishing to write a romance book. Some literary writers--Ron Hansen comes to mind--have more or less made their careers writing historical novels; others--e.g., Madison Smartt Bell--have completely re-made their careers, earning considerably more prestige for their historical novels than any others. Hansen and Bell have no interest in hack work and don't for a second think of themselves as doing such work. Because they aren't. (Read their books if you doubt my word.)

When you think about it, there's no reason why character-driven, realistic fiction need only be situated in contemporary times. You place the exact same character-driven, realistic story a hundred or five hundred years in the past and suddenly it's given the label of historical fiction, even though for the author he's not fundamentally doing anything different from his last novel, set in L.A. in 2007. Oh, for sure, there are some additional concerns. A great deal more research becomes necessary. (But nearly every piece of fiction requires some research.) And if the author is working with an actual figure out of history the author must struggle with the parameters of what "really happened" to this person versus what the author wants to have happen in the novel. These are not unimportant matters, but still at the end of the day the writing process for a literary historical novel is not that much different from writing any literary novel.

Historical fiction is quite the two (or three or six) headed monster, and it appears to be at an interesting crossroads. Tell one reader you're writing a historical novel and you earn a tolerant, even condescending, smile; tell another reader the same thing and you earn awe. Meanwhile, more and more authors give the form a try, discovering its pleasures and considerable challenges. "Form," I just said. Should I have written "genre"? I'm trying not to.

In a coming post: The intriguing case of Caleb Carr. Can an author explore both form and genre in the same novel?

Throwing the F bomb in historical fiction

I blogged last weekend about an issue that arose during a historical fiction section at the recent AWP conference in Denver. In that entry, I took novelist Ron Hansen to task for his rule that a writer of historical fiction should never "knowingly part from fact." Hansen said many things during that session that I do agree with, one of which is that a writer of historical fiction should not use slang or terminology that postdates the era he or she is writing about. That seems a straightforward rule, one that I think most people would find reasonable and even useful. Part of the joy of writing historical fiction, after all, is finding out about previous decades and centuries, including the kinds of phrases used back then. When one is writing a novel set in North America or another English-speaking part of the world, Hansen's rule is easy enough to process and to follow. If Americans didn't say "by the skin of our teeth" in 1791, then you probably don't want to have the characters in your 1790s Boston novel use that phrase.

Things become more complicated, however, when your novel is set in a non-English speaking country or, as in the case of my Van Gogh novel, countries. If you are writing in English for an English speaking audience, then part of the long process of creating your novel is finding the right tone in English for conversations that would have actually occurred in French or German or Dutch or Swahili or whatever. (Assuming, of course, that your made up conversation ever actually happened at all.) So when a word came into parlance in English is less of a pressing matter than whether or not that word effectively renders the flavor of the character's thought. To be specific: At various points in my novel, different characters have reason to express disgust or anger or severe frustration. In a contemporary novel (in English), a character might simply say "F*** it." But can I have the characters in my nineteenth century novel--set in a non-English speaking countries--say this? My answer, so far, has been yes. In fact, the F word might be one of the easier linguisitic issues to put to rest. First, when I use that phrase, or a version of it, (and I don't use it very often) I do it because I think the word fits the emotional tone of the thoughts or conversation depicted. Would Van Gogh have actually used the word "f***"? Probably not. (Although he was fluent in English.) Would Gauguin? Not at all. But that's because they were not native English speakers. In fact, none of what I have them say in the novel they could ever have said--not the precise way I write it--because my characters weren't speaking in English back in Nuenen or Antwerp or Paris or Arles in the 1880s. I realize this may sound like an obvious point, but I make it to emphasize that a writer who is, essentially, trying to translate nineteenth century continental speech into English that a contemporary audience can read and emotionally appreciate, automatically earns a little leeway in terms of the application of Hansen's rule. Besides which, those who have looked into the etymology of the F word--and, by the way, I understand there is a fascinating documentary about this very subject, available through Netflix, titled The F-Bomb: A Documentary--have discovered that in English the word has been used for hundreds of years, well before the dawn of the 1880s. Not only that, but it seems likely this English language slang word is NOT derived from some silly acronym, but from simillar sounding words in Dutch (Van Gogh's native language!), German, Swedish, and Norwegian that mean "to strike" and/or "to copulate." (In fact, the Swedish word for "penis" is fock.) So given that the word existed in English at the same time as the real Van Gogh and the real Gauguin lived in continental Europe I feel all the more justified in using the word for my English-speaking audience--if I feel it accurately expresses the tenor of the characters' thoughts.

Language in a historical fiction is an extremely tricky issue, we can all agree, no matter what we think of Hansen's rule. Some would say, "Isn't the point of any novel to reach and touch the contemporary reading audience?" And "Can any of us really know how people talked and thought day-to-day in a world from 300 or 500 or 1000 years ago?" The answers are Yes and No. Given that, can we just go ahead and use contemporary English, no matter what the era is we're depicting? My answer: Not so fast. The simple fact is that in order to be affecting to the reader the language has to be credible, and in order to be credible it has to feel as if it could belong to the era depicted, even if it finally doesn't. And that's where applying Hansen's rule can be very useful for establishing and maintaining just that credibility.

Parting ways with Hansen

It's been a week since I returned from the AWP conference in Denver, but my head is still buzzing with questions stemming from the very last session I attended, one on historical fiction (the only one on historical fiction), featuring Ron Hansen, Cynthia Mahamdi, and Philip Gerard. It was a good session, one of the best I attended this year, just unfortunately placed in the very last time slot of a three day conference. Ron Hansen is a terrific novelist, someone who never writes a bad book. He's also someone who has a considerable track record in writing crisp, illuminating, perfectly delivered, historical novels. His book Hitler's Niece is one of the finest novels I've ever read in any genre. (I think I might have said that on this blog before.) In the session, Hansen elucidated twelve "rules" for writers of literary historical fiction, and I was gratified to realize that in composing my Van Gogh novel Yellow, I've followed almost every one of them. In fact, most of his rules seemed like good, plain common sense for any conscionable writer aspiring to write something that has value. I like to count myself in that camp. However, I instinctively parted ways with Hansen at one of his rules. Hansen said that the fiction writer should not knowingly part from fact. This is a disservice to the reader, he suggested, who is trusting the writer to render a realistic, even if imagined, picture of the subject. The reader is trusting the writer to get the details right even as the writer is developing his or her story. It is a disservice to both the reader and the subject, Hansen suggested, to put a character in Rio in 1932, for instance, when according to the historical record the character never went near Rio until 1940.

I understand Hansen's point, and I also understand that almost all of the fictive work of historical fiction is not in developing the basic story but in imagining what real historical events looked like, sounded like, smelled like, and felt like for the real participants. To bring alive in scene what is a mere sentence in the historical record. This will mean making arbitrary decisions sometimes as to what someone is wearing, or what diction they use, or how tightly wound they seem, decisions that are virtually identical to the myriad decisions we make when writing a purely imagined, non-historical work of fiction. It's also true that some subjects are so barely sketched in the historical record that a writer has immense leeway in imagining what really did or didn't happen; the writer can wander off in all sorts of interesting directions without "knowingly departing from the facts." (William Styron noted, in his preface to The Confessions of Nat Turner, that it was exactly because so little was known about Nat Turner that he enjoyed taking him up as a subject.) All of this is to say that I understand how a writer can follow Hansen's maxim to the letter and still be free to write a richly felt, deeply imagined work.

But we can never depart from known fact? Never? And for the reason that we might give false ideas to the poor reader, who will mistake our books for biography? Wait a minute. I'm a writer of fiction. That Yellow is novel--not a biography--I have made, and will continue to make, perfectly clear. In fact, it's impossible not to read it without realizing that it sounds nothing like a biography of Van Gogh, just like Hitler's Niece sounds nothing like a biography of Geli Raubal. And if my book is ever published, it will have one of those big, fat, legalistic warnings labels inside proclaiming it to be a work of fiction. So I'm at fault if a reader mistakes it for straight history? No. Sorry. No way. If a reader is reading a novel, the reader should know what the definition of a novel is. If he or she doesn't, that's not the author's fault. Second, and more importantly, sometimes the writer of a historical fiction must depart from fact in order to make his novel work dramatically. There are a couple times I did this in Yellow, and I think of them as some of the most necessary scenes. Without them, I don't know how I could have held my book together as a novel. A novel certainly can't just be a dramatized version of every notable fact in a subject's biography. If you just starting writing ever inch of a person's life you end up with hundreds, if not thousands, of unusable pages. What you have to do is find a way to coordinate a person's life story into a compelling and credible fiction. And fiction--as we all know--counts on having a plot.

I don't want to sound glib here. I understand a commitment to the facts. It was because I was so interested in the specific facts of Van Gogh's life, because those facts so sparked my imagination to action, that I quickly abandoned my original plan, which was to write a novel about a Van Gogh-like painter. I found that so many of the scenes I wanted in the book were so drawn directly from, and depended upon, the man's real life that to use anything else than the his real name would have been ludicrous. And dishonest. And transparent. But by deciding to write a novel about Van Gogh, and not just pen yet another biography of him, I've made a decision that allows me, by definition, to "knowingly part from fact." I will fight to my last breath for my right, as a fiction writer, to do that; and not just my right, but my responsibility too.

Cooking in Paris for Stein

A few times in this blog I've recommended various books of historical fiction, ones I've read during this sabbatical semester or earlier: The True History of the Kelly Gang, Ahab's Wife, Mariette in Ecstasy. (And don't miss an even better historical novel by Ron Hansen, one of the best novels I've ever read in any genre--his Hitler's Niece.) I recently finished a book quite different from any of those and perhaps from any I've read before: The Book of Salt by Monique Truong. Truong's first novel, it is a dreamy, fluid, at times almost formless construction that delves into the mind and the past of a Vietnamese man who worked for several years as the personal chef to Gertrude Stein and Alice B. Toklas. I must admit that it took some time delving into the novel before it completely held me, but Truong's eloquence is irresistable and her pictures of pre-war Paris and of pre-communist Vietnam are intriguing, concrete, believable, and engaging. (Keep an eye out for the appearance of someone who almost certainly is a young Ho Chi Minh.) What you shouldn't anticipate on approaching the book is that it will act as some kind of tell all about Stein and Toklas. Indeed, I venture to guess that more pages in the book are devoted to the protagonist's life before he meets Stein and Toklas than after he does. There's a coolness and even an unexpected lack of curiosity in how he describes the two women; and yet at the same time he seems to know or have intuited some rather intimate details about how they regard and have regarded each other. But, ah hah, this brings me to my last point--the stroke of genius that underlies the book and finally makes sense of nearly every one of Truong's moves. We realize, late in the narrative, that what we are reading was not, in the fictional world of the novel, composed by the chef himself but--similar to The Autobiography of Alice B. Toklas--written by Stein from the perspective of her Vietnamese employee. She apparently has, from a distant remove, regarded him as quite the curiosity over the years, enough to compose a book with him as the narrator. The title of Stein's manuscript? Why, of course: The Book of Salt. This explains so much about the book that I cannot even begin to tell in this brief post. It also means that the Book of Salt is not merely a good read, but will make for a great second and third and fourth read as you uncover more and more gems served up (if you'll excuse the pun) with brilliant and subtle acumen by this very talented writer.

Beginnings of How

In the very first post I wrote for this blog (just click on "About" to see it), I explained how I received the idea to write a novel about Van Gogh in the first place. But that's just one aspect of the origins of this project, the "What." Equally pressing, and this is of course true anytime you write anything, was: "How?" That is, how did I want to approach my subject after I realized that Van Gogh must be my subject? How should I structure the novel? What motifs would I press? I guess you could broadly call these issues of style rather than subject. One very important stylistic decision I made stemmed from having reading Ron Hansen's Mariette in Ecstasy not long before the idea for the novel came to me. If you haven't read Hansen's delicious little book, you must. With exquisite delicacy and much quiet poetry, he details the life of a cloistered nun in upstate New York in the early 20th century. I know, that doesn't sound especially thrilling but, Hansen's book is a wonder and rightly earned considerable acclaim when it appeared in 1991. When I got around to reading it, what I marveled at most was how beautifully contained are each of the chapters. Every chapter, even if it details only single minute in his protagonist's life, is a universe unto itself. Hansen's narrative develops in moment-by moment-by-moment thrusts; there are virtually no bridging sections. This seemed a fascinating and atypical way to structure a novel, but it also felt just "right" for the subject. Since at the time I was exploring the world of monasteries in order to write a novel of my own, I knew how the lives of cloistered nuns and monks are purposefully kept the same--at the surface level--from day to day. They attend services on a fixed schedule, keep to regular work details, and engage in regularly scheduled meals, private prayers, and bible reading. On the surface, it's none too exciting and doesn't seem to ever vary. But that's the point and the purpose. Where the action is--where it's supposed to be--is on the level of the soul: those moments of profound confrontation with yourself or with God, wrapped by external quiet. The structure of Hansen's novel cannily emphasized this philosophy of living.

When I conceived of my Van Gogh novel, I immediately thought of Mariette and wanted to adopt Hansen's structure as my own. For some reason I felt it would be "right" for my subject too. My book would be a small book, composed of a series of wildly divergent but significant moments in the painter's life. Even more so than in a conventional novel, my book would proceed scene by scene, rather than chapter by chapter or year by year. Like Hansen, I would create very few bridges. Well, I've actually stuck to that plan during the four years I've worked on the book. Except the novel, in its present incarnation, is not in the least short. Scenes, after all, require pages to effectively be dramatized, and if you write enough scenes you end up with a heck of a lot of pages. And since, unlike Hansen, I am more or less covering the entire life of my protagonist, brevity is virtually impossible. That said, it's important to recognize that this commitment to scene over summary can sometimes be the most efficient way to cover a broad span of years. I think of some of Jo Ann Beard's essays in her brilliant The Boys of My Youth. She'll present a scene from one year in her life, then jump to another year, and then another, allowing that one scene to represent all the fundamental tensions of the speaker during that period. In this way, she can suggest her whole life in 30 or 40 pages, whereas as a more overtly inclusive narrative would have taken many more pages and perhaps not achieved any greater payoff. As I said: efficient. I guess with my novel so far I've not been courgeous enough to let a single scene represent Vincent in each of his various life "periods." (He had several such periods.) Some of these periods--notably Paris and Arles--I've rendered in several scenes. At this point, I've written so many scenes I have a hefty--too hefty?--novel, even after doing quite a lot of cutting. But the thing is, I can honestly say that each of the scenes contributes something important and unique to my picture of Van Gogh. I'm glad I developed the novel in this manner and can't really imagine doing it any other way. I only hope my book can do justice to the influence of Hansen's beautiful and exotic Mariette.