Reflections on a Dragon Tattoo

My feelings about The Girl With the Dragon Tattoo are a matter of record. No other cultural sensation of recent years has left me so cold: where others see a masterpiece, I see a book that is amateurishly plotted, lurid but airless, overlong, and, worst of all, often grindingly dull. I’m not passing judgment on the novel’s many fans; only trying, unsuccessfully, to figure out what they find so compelling. I’m on the outside, looking in. Which made me all the more interested, paradoxically, in seeing David Fincher’s film of the book. As I’ve said before, with Zodiac and The Social Network, Fincher has gone from a filmmaker toward whom I’ve always felt considerable ambivalence to one of my four or five favorite contemporary directors. He’s an impeccable craftsman with a nice, chilly style, and to my eyes, he seemed like just the man to pare away the worst of the book’s shortcomings to reveal the germ of a decent story at its heart.

The good news is that the movie is much better than the book. Fincher and screenwriter Steven Zaillian skillfully foreshorten the novel’s interminable opening and closing sections, cut down on the number of meaningless suspects, and make the logic of the investigation, if not exactly plausible, at least visually comprehensible. In many ways, this is a more impressive display of Fincher’s craft than a more engaging story might have afforded: for a movie with little conventional suspense and even less real action, it’s surprisingly absorbing, and seems much shorter than its actual length of nearly three hours. Like all of Fincher’s movies, it looks and sounds great. And the cast is excellent, especially Rooney Mara as Lisbeth: it’s a performance based as much on makeup and costume design as any real conception of the character—much of her acting is done by the back of her head, and those amazing earrings—but Mara commits fearlessly to the part, and whenever she’s onscreen, the movie gains an additional charge.

Unfortunately, while the film does a nice job of addressing the story’s tedium, it doesn’t do much for its essential pointlessness. What, exactly, is this movie about? Like the book, the film ruminates endlessly on the complexities of the Vanger company and its tangled family tree, only to give us a killer at the end whose identity and motivations are completely arbitrary. The characters make wildly implausible deductions and even more inexplicable decisions, as when Blomkvist, effectively portrayed by Daniel Craig, figures out who the killer is, then rushes over immediately to the suspect’s isolated house, alone and unarmed. Perhaps most unforgivably, while the movie, like the book, is superficially concerned with violence against women, it has nothing interesting to say on the subject—aside from endorsing some astonishing forms of revenge—and often seems content to simply titillate the audience. From Fincher, who is capable of much better things, this is a particular disappointment.

Despite its obvious technical merits, then, The Girl With the Dragon Tattoo feels like a step backward for one of our most interesting directors. If Fincher had only made Seven, The Game, and Panic Room, I might have felt differently: in that case, this film would have seemed like the best he was capable of delivering. But after The Social Network, and in particular after Zodiac, Fincher has emerged as a director who can follow through masterfully on genre conventions while also teasing out deeper possibilities. He’s still a master of mise-en-scène, and, like Hitchcock, he’s fond of nice sick touches—his use of “Orinoco Flow (Sail Away)” is especially inspired—but Dragon Tattoo finds him oddly unengaged. As the credits roll, we know we’ve been treated to a sleek, professional studio product, with isolated flashes of beauty and cruelty, but we aren’t sure why. And I don’t think Fincher knows, either.

The Fugitive and the art of beginnings

The other day, as we were talking about the divergent career paths that the leads of Star Wars had taken, my wife asked me what the last great Harrison Ford movie had been. I answered without hesitation: The Fugitive. And, immediately, I wanted to watch it again. Much to my relief, I found that it’s still a great movie. In particular, the first half hour strikes me as close to perfect: it plunges us right into the action, elegantly introduces the hero and his dilemma, and then all but throws us into the next stage of the story. Ideally, on first viewing, we’re too caught up in the narrative to think about the craft on display. It might even seem easy. But it isn’t.

Which brings us to a larger question: at what point in the story should a novel or movie begin? If the answer seems obvious—a story should begin at the beginning—that’s a good thing, because it means we’ve been spoiled by works of art that, by and large, begin at the right time. But the question of where an extended narrative should begin is as old as the Iliad and as recent as The Girl With the Dragon Tattoo. (I don’t want to harp on this subject yet again, but if Stieg Larsson had known exactly where to begin and end his story, that book would have been infinitely more readable.)

The short answer is that the narrative should begin as late in the story as possible. In movie terms: burn the first reel. David Mamet, as always, is endlessly quotable:

Almost any film can be improved by throwing out the first ten minutes. That exposition, which assuaged the script reader, the coverage writer, the studio exec, the star and her handlers puts the audience to sleep sleep sleep. Get right into the action, and the audience will figure it out. (Simple test, for the unbelieving: when you walk into a bar and see a drama on the television, you’ve missed the exposition. Do you have any trouble whatever understanding what’s going on?)

And this is as true for novels as of movies, if not more so. One useful test: on rereading a novel, do you skip the first thirty pages to get to the good stuff? If so, make a careful note of where you begin rereading, because that’s more or less where the novel should have begun. The same principle applies if you leave off reading before the end. For instance, I rarely reread the opening of The Day of the Jackal, and I’ll usually skip several of the explanatory chapters near the end of The Silence of the Lambs. And these are two beautifully constructed novels, which implies how hard it can be to put together the pieces.

In the case of The Fugitive, the credited screenwriters, Jeb Stuart and David Twohy, the director, Andrew Davis, and the six editors made a series of strong choices. (Perhaps luck played a role as well: filming was evidently begun before the script was finished, and the screenplay had a lot of uncredited hands.) The film could have opened with an ordinary day in the life of Dr. Richard Kimble, or at the party in which he and his wife were last seen, or even at his graduation from medical school. Instead, it opens exactly where the real story begins: at the moment of his wife’s murder. Necessary information is conveyed in a series of rapid flashbacks. And Kimble is arrested, tried, and convicted before the credits are over. (After such a virtuoso opening, it’s no surprise that the movie’s second half is a little deflating.)

Of course, if your movie is called The Fugitive, and based on a famous television show of the same name, you probably have a pretty good sense of where your story needs to start. For an original novel, it isn’t always as clear. In general, as John Gardner says, a novel should open “when the action actually begins,” which comes perilously close to tautology. Ultimately, experience is the only guide. At the beginning, it’s likely that the author will write one or more opening chapters that will need to be cut, later on, as the true shape of the novel becomes clear. Which is fine. But the best solution, by far, is not to write the unnecessary scenes in the first place.

(That said, I’m not a fan of novels or movies that begin at a dramatic moment near the climax, then flash back to show how the protagonist got into this mess. There are exceptions, of course—The Usual Suspects is one of the greatest, and Michael Clayton just barely gets away with it—but for the most part, it makes the story look, as Gardner puts it, “gimmicky and self-regarding.” Far better, I think, to find a striking scene that takes place early in the story’s chronology, and begin there. Every shift in time forces the reader to stop and regroup. The novel will be more readable if you pick the right opening moment and run with it.)

The Girl Whose Books Aren’t Very Good

At first glance, Joan Acocella’s openly contemptuous piece on Stieg Larsson in The New Yorker looks like the comprehensive takedown that the Millennium trilogy has long deserved, but I don’t think she quite pulls it off. Acocella devotes a few paragraphs to the trilogy’s “almost comical faults”—bad writing, poor dialogue, irrelevant detail—but concludes by saying that Larsson is “a very good storyteller.” Which is almost exactly wrong. I’ve only read The Girl With The Dragon Tattoo, and I have no plans to read the others, but it seems clear to me that Larsson’s faults—and virtues—are very different from the ones that Acocella describes.

First, the good. I don’t think that Larsson’s writing is nearly as bad as others have said. Although it’s true that almost every paragraph could be revised to be sharper and tighter—as editor June Casagrande has done, to amusing effect—that’s true of virtually all mainstream suspense fiction, and Larsson is no worse an offender than most. If nothing else, his prose was smooth and propulsive enough to keep me reading Dragon Tattoo for what must have been hundreds of pages, even as I slowly realized that nothing of interest was happening, or going to happen. (More about this later.)

What else? Larsson knows how to create an atmospheric setting, even if he rarely follows through. And his two main protagonists are perfectly fine. Blomkvist is something of a Mary Sue, yes, but again, he’s no worse than many other lead characters in suspense fiction. And Lisbeth Salander, at her best, is pretty much as advertised: an intriguing, memorable avenging angel. Her scene of revenge against her abusive guardian is the only really good scene in the entire first novel, and it’s so powerful that it almost provides enough momentum to propel the reader through the ensuing three hundred pages of complete inaction. Almost.

And that’s the problem. Despite what Acocella says, I don’t think that Larsson is a good storyteller at all. The Girl With the Dragon Tattoo presents itself as a locked room mystery, but the solution is so banal that I had to read it twice to make sure I hadn’t missed anything more interesting. Deductions are made by such farfetched devices as a girl’s facial expression in an old photograph, which, even if it were plausible, just doesn’t work on the page. And the author’s idea of compounding the mystery is to introduce us to dozens of interchangeable uncles, aunts, and cousins, all potential suspects, none of them memorable. As Borges wrote:

The amateurs [of the detective story]…are partial to the story of a jewel placed within the reach of fifteen men—that is, of fifteen names, because we know nothing about their characters—which then disappears into the heavy fist of one of them. They imagine that the act of ascertaining to which name the fist belongs is of considerable interest.

And even these shortcomings wouldn’t matter as much if the novel were at least streamlined and concise, which it isn’t. The ideal thriller should be a perfect machine with no superfluous parts, but Dragon Tattoo is so padded that it resembles another Borges creation, the Book of Sand, in which the reader is always an infinite number of pages from the beginning and the end. It’s no exaggeration to say that, given half an hour and a red pencil, a third of the novel could easily have been cut, especially from its opening and closing sections, with no loss whatsoever.

So why are these novels so popular? Acocella seems to have no idea, aside from the possibility that readers are turned on by the trilogy’s lurid combination of feminism and rape, or, in a particularly lame conjecture, by the books’ “up-to-dateness, particularly of the technological variety.” The real answer, I suspect, is that Larsson’s combination of superficial readability and intense boredom has convinced a lot of readers that they’re reading something good for them. And for those who think I’m being unfair, I have a simple test. Read The Silence of The Lambs. Then read Dragon Tattoo again. And if you still feel like defending Stieg Larsson, then, perhaps, we can talk.