Girl on Film: Kubrick’s Lolita versus Lyne’s
This is the second in my ongoing series of posts on Nabokov’s Lolita, which I’m “teaching” over a twelve-week “class” that started yesterday. The first installment in the series is here.
There’s a lot to be learned about a novel from seeing it in media other than the accustomed one, for instance, reading it in translation or watching a movie adaptation. A novel by design leaves a great deal to the imagination, and it’s always interesting to see how someone else has imagined it, particularly when that someone has to imagine it down to the details.
To the best of my knowledge, there are only two movie adaptations of Lolita in English, Stanley Kubrick’s 1962 version, and Adrian Lyne’s 1997 one. The two films involve radically different takes on Lolita, and reactions to them diverge in radically different ways. Kubrick gives us a version of Lolita as dark comedy; Lyne gives us one as tragedy. The general consensus is that Kubrick captures the comedy of the novel, sacrificing textual fidelity in the process. Meanwhile, Lyne is more faithful to the text of Lolita, in the prosaic sense of conveying the plot and reproducing the dialogue, but filters out the comedy, an essential omission. On this view, Kubrick gets the letter of the text wrong but the spirit right; Lyne gets the letter right but the spirit wrong. If the spirit matters more than the letter, Kubrick wins. On the reverse assumptions, Lyne does.
I reject this consensus. As far as I’m concerned, Kubrick’s Lolita is an excruciating, unwatchable car wreck of a film. It’s not that it gets the spirit right but the letter wrong, but that (as far as I’m concerned) it gets everything wrong. Three of the major characters–Humbert, Charlotte, and Lolita–are (as I see it) miscast, and all three actors, in my view, mis-play their character. A fourth character, Quilty, is supposed to remain in the background, but migrates to the foreground and is allowed to steal the show. Even conceding the need to invent some dialogue not in the book, far too much is invented, much of it bad, or at least bad by comparison with what we get in the novel. Since much of this dialogue is delivered by Quilty, what we end up getting is a lot of bad dialogue improperly foregrounded by a minor character.
The selling point of the film is supposed to be its humor, but I didn’t find it nearly as funny as many viewers have. I chuckled a couple of times, to be sure–Quilty is played by Peter Sellers–but there really wasn’t much to laugh at. People sometimes say that Kubrick was “unfaithful” to the plot, which is true, I guess, if you can be “unfaithful” to something by ignoring its existence. I would put things less nicely. Huge swatches of the plot are simply missing in Kubrick’s Lolita, creating a series of unresolved mysteries in the viewer’s mind about what is happening on the screen. Finally, while many people have tried to suggest that the novel isn’t “really” about sex, Kubrick’s delicacy about its sexual content (partly imposed on him by censorship) flout the text in an obvious way. There are a lot of handjobs in Lolita, and you can’t convey that by discreetly writing handjobs out of existence.
As for Lyne, contra common opinion, I don’t think he merely got the letter of the novel right in some boringly pedestrian way, doing away with its ineffable spirit. I would say that he got the whole package right: his Lolita is a masterpiece in both letter and spirit. It gets everything right about the book that Kubrick gets wrong. But I don’t so much want to pursue or prove this claim as make a more basic suggestion about the debate as a whole, one more general than these films, and in fact, more general than Lolita.
Lolita depicts an act or series of acts of child rape carried out over several years. Lolita is about twelve when her “affair” with Humbert begins, and remains in his custody, by coercion, until the age of fourteen or fifteen. There’s no way of bypassing this basic fact, and no reason to do so. In depicting Lolita on film, a basic question one has to ask is whether the depiction of such a sequence could in principle be fodder for comedy of any kind, “black” or otherwise, or must of necessity be tragic. In my view, the question has to be answered, at least provisionally, before an extended encounter with the novel. I don’t mean that one’s answer can’t change or develop along the way. What I mean is that engagement is significantly enhanced by clarifying the issue from the outset. Can an extended act of child rape conceivably be the stuff of comedy? Or is that possibility simply inconceivable, hence impossible, at least practically so?
You could, I suppose, punt on the whole question, and insist that Lolita is sui generis, neither comedy nor tragedy of any kind. I won’t dispute that here. I don’t want to insist that “comedy” and “tragedy” are exhaustive of the universe of possibilities, but I do want to insist that they’re exclusive options. If something is a comedy, it isn’t a tragedy, and vice versa. Given that: suppose ex hypothesi that you’re forced to choose between exclusive categories for categorizing Lolita, comedy or tragedy, bracketing the others. Which one applies here? Which one would it have to be? Which one would you say it had to be even if you hadn’t yet read the novel?
No matter how many semantic games you play with the concept, a “comedy” in the contemporary sense is funny, and meant to be funny. You can add as many adjectives to “comedy” as you like, e.g., “dark,” or “black,” or whatever. It doesn’t change things all that much. A comedy is funny. But an extended act of child rape isn’t funny. Since Lolita depicts an extended act of child rape, it isn’t fundamentally a work of humor, even if parts of it have an undeniably mordant humor about them. So it’s not a comedy. Philistine, I realize. But undeniable.
In order to play Lolita for laughs, you have to do one of two things. Either you have to ignore the fact that the book is about child rape, or you have to think that a depiction of child rape can provide the stuff of humor. But neither of these options is very plausible, and neither is what Nabokov actually does. It is true that there is more to Lolita than child rape, but there’s no denying that child rape is what the novel is about in some basic way. It’s true that humor might arise in the interstices of a depiction of child rape, but that’s not the same as saying that such a depiction makes the work a comedy.
Consider an analogy. Technically, Shakespeare’s Merchant of Venice is a “comedy,” at least in the sense that it doesn’t meet the formal criteria of a Shakespearean tragedy. But setting that technical issue aside, there is only one way to treat The Merchant of Venice as comic in the colloquial sense of being funny: you have to treat Shylock as an object of derision.
Why? Well, the derision directed at Shylock in the play is almost entirely anti-Semitic. To treat Shylock as an object of derision, then, you would have to own this anti-Semitism yourself; you’d have to treat the specifically anti-Semitic derision aimed at Shylock as authentically funny. To do this, I think you would have to find the anti-Semitism itself funny. But very marginal cases aside–e.g., in some very narrow context within a stand-up routine where anti-Semitism is played for laughs and delivered by a Jew, but kept within very circumscribed limits–anti-Semitism isn’t funny. And though I’m being cautious in allowing for some “very marginal cases,” I’m actually pretty skeptical that these cases really exist. I doubt that anti-Semitism is funny even in “very marginal cases.”
In short, the only feasible way to play The Merchant of Venice as a comedy is to be a wholehearted anti-Semite, i.e., first-personally to adopt a wholehearted anti-Semitism, finding it hilarious that Shylock comes in for such demeaning derision as we find in the play. If you reject anti-Semitism, by contrast, it’s psychologically impossible to find humor in the way Shylock is treated.
I don’t dispute that a comic interpretation of The Merchant of Venice is both psychologically and dramatically possible; indeed, it’s the way Merchant of Venice was played for a long time. But it’s only possible to confirmed anti-Semites. A dramatist who takes that route suborns sympathy for the very worst things depicted in the work, and turns the work itself into a dramatic cesspool. The Merchant of Venice is not, in my view, an inherently anti-Semitic play, but if you play it for laughs, it inescapably becomes one.
The only way to avoid the descent into anti-Semitism is to treat Merchant as a tragedy. Doing so, I should add, is not misplaced piety or a problematic exercise in political correctness. The play genuinely lends itself to a tragic depiction: though technically a comedy, Merchant of Venice is substantively tragic; Shylock, almost the center of the play, is a fundamentally tragic figure.
In re-imagining The Merchant of Venice in this way in his 2004 film version, Michael Radford didn’t distort the play for some woke end extrinsic to the play. He made a choice about the genre of the play–tragedy–and filmed it accordingly. It’s true that he did so by invoking a moral commitment not literally in the play, most obviously that anti-Semitism is wrong. But you’d end up doing the same thing from the reverse end if you treated the play as a comedy.
A presupposition of the belief that the expression of anti-Semitism is uproariously funny is the belief that there’s nothing wrong with it. That’s not literally in the play, either. No explicitly moral belief or principle is. There is no stage direction prior to Act I, Scene 1 that says, “Presuppose the truth, falsity, rightness, or wrongness of anti-Semitism. Then proceed with the play.” You can take the play either way, but you have to take the play one way or the other. Each potentiality is there, to put the point in Aristotelian terms. The choice between them is yours.
Same with Lolita. In my view, Lyne correctly plays Lolita as a tragedy. I can’t tell you whether he literally did so by clarifying all of his background beliefs well before he got to the work of filming, or it came through the screenplay, or what. It doesn’t matter. Either way, he basically got things right: the aesthetic quality of the production is in part a function of Lyne’s moral and conceptual clarity about the content of the novel.
In doing this, Lyne filters out some (not all) of the humor in the book. That is, I guess, something of a loss, but it’s not nearly as big a loss as his critics have suggested. No film can capture everything in a great novel; Nabokov’s novel exhausts what any film can depict of it, in just the way, and for just the reason, that skylarks exhaust what poets like Shelley have tried to say about them. Ultimately, something has to go. If the novel is a tragedy, then it’s acceptable for some of the humor to go. The humor is important, but still subordinate to this more basic feature of the book, its tragic nature.
What Kubrick does is the reverse. He treats Lolita as a comedy and plays it for laughs, omitting and/or butchering its tragic dimensions. This is a much bigger problem than anything in Lyne’s film. It gets the aesthetic priorities wrong, sanitizes the sexually coercive nature of the Humbert-Lolita relationship, and leaves us with a basically unfunny and unintelligible series of gags, played out over two excruciating hours.
The result is an object lesson on how not to film Lolita. It’s also an object lesson, in reverse, on how to think about Lolita, both novel and film, and how to think about the relationship between art and ethics generally. The basic guidepost is that the connection between the two things is not straightforward but still there–hard to articulate, but impossible to explain away. Watch both films after reading the novel, and I think you’ll see what I mean.
Thanks to Charles Persky, Gail Persky, and Hilary Persky for endless hours of discussion on The Merchant of Venice, and to Scott Fischbein for the offhand comment about Shelley’s “To a Skylark” that made it into the post. None of these people can be held responsible for anything I say.
