Posts Tagged ‘The Third Man’
Comedy, as we all know intuitively, is largely built on threes. It often shows the same thing three times with slight variations, followed by a kicker at the end, which is why so many jokes are built around three different nationalities, religions, or professions, like those about the mathematician, the physicist, and the engineer. There’s the famous comedy triple, in which two items set up a pattern, followed by a third that serves as a punchline. (There are countless examples, but I’ve always liked this one from The Simpsons: “Well, little girl, I’ve had a lot of jobs in my day: whale hunter, seal clubber, president of the Fox network…”) A similar rule applies to magic, which depends on the basic pattern of setup, development, and surprise climax. In Magic and Showmanship, Henning Nelms describes a trick in which a color-changing fan is used to magically dye handkerchiefs different colors, and then says:
Commercial color-changing fans can display four different hues. But this is bad showmanship. Dyeing one is trivial. Dyeing two arouses interest. Dyeing three provides your climax. There is no reason to add an anticlimax simply because you are prepared to do so.
So why is the number three so powerful? For the same reason that one point is just a point, two points is a line, and three points, suddenly, is structure. Our brains are wired to look for patterns, and it takes three items to confirm or deny that a pattern exists—and it can be very satisfying either to be given the payoff we’ve been expecting or to be shown how cleverly we’ve been misled. Writing about The Godfather, David Thomson speaks of “the sinister charm of action foreseen, spelled out, and finally delivered,” as when Michael kills Sollozzo and McCloskey. “It is a killing in which we are his accomplices,” Thomson says, and three is the minimum number of story points required for the reader to actively conspire in the narrative. This is why most of our stories, from jokes to fairy tales to novels, still consist of a beginning, a middle, and an end. (Or, as Philip Larkin puts it, “a beginning, a muddle, and an end.”)
This also applies to a story’s constituent parts. Narratives tend to have a sort of fractal structure: an individual chapter or scene will often have the same three-act structure as the story as a whole. This often applies to the movie scenes we tend to remember most vividly, which are structured as miniature plays—think of Holly’s first meeting with Harry in The Third Man. My own novels and stories are usually structured in three acts, to the point where I use numbered sections even in short novelettes, and that applies to individual chapters as well. When I’m outlining a chapter, I’m generally thinking in threes, even before I know what will happen: I’ve learned from experience that three story beats is a strong foundation on which to build a chapter, for the same reason that a tripod needs three legs to stand, so I always make sure that the chapter falls into three roughly similar parts, at least in the first draft.
And yet here’s the funny thing: when it comes to the final draft of a chapter, the first and third parts often don’t need to be there. I’ve spoken before about the importance of writing the middle—that is, of cutting the opening and closing sections of a chapter and jumping from the middle of one scene to the next—and I’ve often noticed that rough drafts spend too much time moving toward and away from the real center of interest. In short, the rule of three is invaluable for structuring a first draft, but in the final version, much of it can be thrown away. In my experience, it’s best to reserve the full three-act treatment for big, climactic scenes, while for transitional chapters or sequences, usually only the middle is necessary. The reader can fill in the first and last parts on his or her own—but only if they’ve been written and cut in the first place. They’re still there, but they’re invisible. And that’s how you use the rule of three.
Even for passionate movie lovers, two things tend to date the classic films of the thirties and forties: their sets, with the inescapable smell of the studio, and their orchestral scores, which to modern ears tend to sound depressingly alike. It’s quite possible, then, that we have both the city of Vienna and Anton Karas to thank for the fact that The Third Man still seems so fresh. The zither score, combined with the extraordinary locations, result in a film that seems both utterly of its time and completely modern—it requires less of a mental adjustment to enjoy than any other movie of its era I know. Combine this with Graham Greene’s great script, with its uncredited contributions from Orson Welles and others, and we have what is both the breeziest and darkest of noirs, a film I love so much that I steal from it directly both in my novelette “Kawataro” and the conclusion of my novel City of Exiles.
Everyone knows how completely Welles dominates the movie with only a reel or so of screen time—which, while delicious, seems much more of its period than the rest of the film—to the point where our memory of Harry Lime tends to overshadow the rest of the cast: Joseph Cotten, the very moving Alida Valli, and especially Trevor Howard as Major Calloway, who contributes perhaps the film’s most stylish performance. The big moments—Harry’s entrance, the ferris wheel scene, the great closing shot—are deservedly famous, but I also like the small touches: the wizened little boy with the ball; the moment when Sgt. Paine (the wonderful Bernard Lee) loads the picture of a rhinoceros into the slide projector by mistake; or the glimpses we get into the work of hack writer Holly Martens though the eyes of his admiring readers: “I never knew there were snake charmers in Texas.” But as Carol Reed’s great film reminds us, there are certainly snakes in Vienna—and very charming ones at that.
On Monday: Kubrick, of course, but not the one you were expecting.
In his nice little book Writing the Novel, Lawrence Block, while describing how he incorporates all kinds of disparate elements into his fiction, uses an image for the creative process that I’ve always thought was particularly appropriate:
I may borrow a bit of physical description, for example, or a mannerism, or an oddity of speech. I may take an incident in the life of someone I know and use it as an item of background data in the life of one of my characters. Little touches of this sort get threaded into my characters much as bits of ribbon and cloth are woven into a songbird’s nest—for color, to tighten things up, and because they caught my eye and seemed to belong there.
Most writers, I imagine, can relate to this. As carefully as any novel or story may be planned, many of its constituent parts will end up being the result of chance, impulse, or random inspiration. “Kawataro” is no exception. Although what I’ve described so far might sound like a fairly rational process, that rationality, if it exists at all, occurs mostly in the intermediate planning stage. When it comes to the details of the novel itself—the characters, the scenes, the small touches that make a story live—the process is much more intuitive, and the results can take even the author by surprise.
The backgrounds of the characters in “Kawataro,” for instance, were a combination of pragmatism and personal inclination. For my viewpoint character, Hakaru, I had a particular type in mind: a smart, observant outsider, but not a scientist, which would allow me to explain certain concepts to the reader in a way that was hopefully unobtrusive. I’ve used the figure of a journalist in a number of stories (including the upcoming “Warning Sign” and “The Boneless One”), partly because I’m married to one, but also because it’s a job that involves asking questions and going into unusual places, which is useful from a storytelling point of view. For a change of pace, I decided to have Hakaru (named, incidentally, for this man) be a videographer with a research background. I knew that projects like the one I was describing were usually videotaped, so he had a good reason for being there. Plus I’ve done a lot of video production myself, so I could easily describe his work if necessary (although it ended up not entering the story at all).
My other main character, Dr. Nakaya, was a bit more determined by the plot I had already sketched out. She had to be a scientist involved in the study of language formation among the burakumin of my fictional village. At some point, it occurred to me that she might also be a burakumin herself. Once these details had been established, her character quickly fell into place: intelligent, slightly severe, but emotionally involved with the predicament of these villagers in ways that were only gradually revealed. As for the other characters, they were mostly functional types—a few fell into the category of characters, familiar from The X-Files, destined only to be victims—but I tried to invest them with at least some specificity. (For some reason, I love Miyamoto’s pink shirt, which is inspired by a similar shirt worn by a figure in The Cove.) And the three sinister children at the heart of the story were clearly rooted in my memories of spooky kids from The Grudge and similar movies, with one of them wearing a red raincoat that was my homage to Don’t Look Now. (It’s an homage that would seem overly obvious in a straight horror movie, but which works pretty well in a different genre.)
Now that I had a general plot and a cast of characters, all that remained was to fill out the story itself. Many of the scenes were dictated by the shape of the conventional story I’d chosen: an outsider arrives in a small town, meets the locals, is confronted with violent and seemingly supernatural events, and finally discovers a rational explanation. In the details, though, I was free to indulge myself. The scene in which a little girl with a bouncing ball watches Dr. Nakaya argue with Miyamoto, then later implicates her in his murder, was a straight homage to The Third Man. Many of the visual details of the story—the rain, the figure in the woods, the children’s drawings unexpectedly revealing a monster—were taken from the vocabulary of horror movies. The layout of my imaginary village determined the beats of the chase scenes. And the image of the dead innkeeper, folded up like a frog, came from a dream I had over ten years ago, which I was glad to finally use here.
In the end, then, I had a story constructed from many dissimilar elements—an article in a science magazine, a Japanese legend, a few character ideas, memories of favorite movies, even dreams—which all came together, I hope, in a seamless and inevitable way. Tomorrow, I’ll wind up the discussion by talking a bit about the revision and submission process, and how I feel about the story that resulted. (For other installments in this series, please see here, here, and here.)
Yesterday I got back from my trip to London, where I spent a week looking at locations for Midrash, the sequel to Kamera. For just over six days, I lurked around neighborhoods like Shoreditch, Holland Park, Stoke Newington, and Golders Green; studied landmarks like the Olympia Exhibition Centre and the Old Bailey; and even indulged in a six-hour side trip to Brussels. I kept good notes, took a lot of pictures, and seriously destroyed my feet—next time, I’m bringing better shoes. And I came away not only with a substantial trove of information for my novel, but also some reflections on the role of location research in the writing process itself.
At first glance, it might seem that direct experience of a novel’s setting is essential, especially for a story supposedly based on careful research. A location contains crucial information—sights, sounds, smells, and human interactions—that can’t be acquired in any other way: I know from experience that an hour in Bombay will teach you things about India that you’d never learn from a lifetime of reading. And there’s little doubt that a novel would benefit from what Werner Herzog, according to Roger Ebert, calls “the voodoo of location” in movies—the idea that locations “seep into performances and photography and give a special texture to the film.”
Yet the issue isn’t quite as straightforward as it seems. Atmosphere is no substitute for story, and excessive use of location research can burden a novel with inessential detail, as we sometimes see in late Michener. And many good or great books have been written without the benefit of actual travel. Saul Bellow wrote Henderson the Rain King without going to Africa, at least as far as I know, and more recently, Scott Smith produced the very good horror novel The Ruins without setting foot in Mexico, although it couldn’t have been hard to make the trip. And the number of classic films not shot on location is impossible to count—after all, nobody on Casablanca got anywhere close to Morocco. (Although it’s hard to imagine The Third Man being shot anywhere but Vienna itself.)
For both movies and novels, the “truth” of a location lies between reality and illusion. No matter how heavily researched a novel’s setting may be, there will always be rooms, houses, and streets constructed entirely from the author’s imagination. The same is all the more true for film, where even the most convincing locations often turn out to be made of spit and cardboard. Some of my favorite cinematic locations are from Powell and Pressburger’s I Know Where I’m Going!, which makes extraordinary use of the Inner Hebrides. Yet the movie’s male lead, Roger Livesey, never came close to Scotland: he filmed all of his scenes in the studio, with a double for long shots, and the movie often cuts between set and location from one angle to the next.
What matters, in the end, is the work itself. As I’ve noted elsewhere about other kinds of research, location work isn’t about factual accuracy, but about furnishing the imagination. The author’s inner eye can play quite profitably in the locations where the novel itself will take place—for Kamera, I spent many happy days haunting the boardwalks of Brighton Beach—but there’s also ample material for dreams in the pages of an atlas, especially when it’s out of date. Sooner or later, at some point in the process, real locations fall away, leaving only what remains on the page. And as much as I loved my trip to London, I’m also aware that it’s only now, back at my desk, that the real location work can begin.
Any list of favorite movies—much less one of favorite screenplays, where the writer’s contribution can be so hard to separate from that of the director and editor—ends up being more about the compiler than anything else. My own list betrays a personal fondness for dense, complicated stories over quiet simplicity, which is arguably the harder of the two to pull off. All in all, though, I’ll stand by these choices—though I’m somewhat surprised to see that one of my top films stars Kevin Spacey, another stars Gabriel Byrne, and another, perhaps inevitably, stars both:
1. Seven Samurai. As far as I’m concerned, this the greatest screen story of all time—a massively detailed film of more than three hours that establishes its central conflict in the first minute, involves us in the lives of more than a dozen important characters, and treats us to the immense satisfaction of seeing epic action foreshadowed, spelled out, and unforgettably delivered. Writers: Akira Kurosawa, Shinobu Hashimoto, Hideo Oguni.
2. L.A. Confidential. A script so good that it forever fooled me into thinking that there was a place in Hollywood for layered, complicated stories, saturated with ideas and atmosphere, with three central characters but no obvious hero. Well, there isn’t. But watching this movie makes you almost believe otherwise. Writers: Brian Helgeland and Curtis Hanson, based on a novel by James Ellroy.
3. The Red Shoes. All of Powell and Pressburger’s screenplays are amazing, but this is the one that fills me with the most awe. Like L.A. Confidential, it effortlessly establishes three major characters—and many minor ones—while ushering us into a world that seems both strange and familiar, with a range of tones that spans realism, surrealism, melodrama, and, in the end, merciless tragedy. Writers: Michael Powell and Emeric Pressburger.
4. The Usual Suspects. The closest thing I’ve seen to a perfect clockwork screenplay, layered with small visual and verbal delights in every scene, all leading up to that famous closing surprise (which makes increasingly less sense to me as time goes on). To quote theater critic Walter Kerr, The Usual Suspects is a watch that laughs—and there’s a hell of a cuckoo inside. Writer: Christopher McQuarrie (though many of the best moments, including the closing montage of dialogue, were created in the editing room).
5. Casablanca. The first forty minutes, in particular, are the best I’ve seen in any movie, in terms of serenely establishing character, location, and conflict in a way that seems as natural as wandering into Rick’s Place out of the hot desert night. The second act has a few narrative lumps—I’m not a fan of flashbacks in general, even when they feature Bogart and Bergman in Paris—but as for the finale, well, nothing more needs to be said. Writers: Julius J. Epstein, Philip G. Epstein, Howard Koch, based on the play Everybody Comes to Rick’s.
6. Miller’s Crossing. It took me years to warm up to this movie, but now that I know it inside and out, I can only marvel at how beautifully all the pieces fit, even if the writers evidently made it up as they went along. (They wrote Barton Fink, on a break, while trying to figure out how to resolve the plot.) It’s still the last of the great color noirs. Writers: Joel and Ethan Coen.
7. The Last Temptation of Christ. I was going to put Taxi Driver here, but this is really Schrader’s—and Scorsese’s—masterpiece: marvelously structured, moving, and more intelligent than so deeply religious a movie has any right to be. The last half hour rarely fails to bring me to tears, though never at the same place twice. Writer: Paul Schrader, based on a novel by Nikos Kazantzakis.
8. The Third Man. The perfect blend of plot, location, and atmosphere, sinister yet romantic, with grotesque supporting characters lurking in the ruins like gargoyles. It all builds to that heartbreaking final image—the greatest closing shot in the history of movies—which wasn’t in the original script at all. Writer: Graham Greene (though Orson Welles wrote his own speech about the cuckoo clocks).
9. Psycho. Yes, yes, the closing psychiatrist’s speech is terrible. But up until that final moment, it’s perfectly structured and paced, with the greatest narrative fake-out of all time—one that works so well that I’m still faintly shocked, whenever I first see the Bates Motel sign, at remembering which movie I’m really watching. Writer: Joseph Stefano, based on a novel by Robert Bloch.
10. Star Trek II: The Wrath of Khan. Pauline Kael called it “endlessly inventive,” and it is, cobbling together a plot, as I’ve described elsewhere, from six different screenplay drafts and a random handful of science fiction elements, and having it all seem relaxed, witty, and inevitable. Writers: Credited to Harve Bennett and Jack B. Sowards, but really Nicholas Meyer.
Honorable mention: Citizen Kane, The Godfather, Chinatown, Raiders of the Lost Ark, Blue Velvet, A Hard Day’s Night, The Treasure of the Sierra Madre, and many others on the definitive Writer’s Guild list.
Warning: Visual spoilers follow. Cover your eyes!
As I’ve noted before, the last line of a novel is almost always of interest, but the last line of a movie generally isn’t. It isn’t hard to understand why: movies are primarily a visual medium, after all, and there’s a sense in which even the most brilliant dialogue can often seem beside the point. And as much the writer in me wants to believe otherwise, audiences don’t go to the movies to listen to words: they go to look at pictures.
Perhaps inevitably, then, there are significantly more great closing shots in film than there are great curtain lines. Indeed, the last shot of nearly every great film is memorable, so the list of finalists can easily expand into the dozens. Here, though, in no particular order, are twelve of my favorites. Click or mouse over for the titles: