Jim Emerson has done a nifty video essay on the car chase sequence from The Dark Knight, articulating with the film’s own images why he feels that Nolan’s action choreography — more specifically, his framing and editing — adds up to a jumbled, incoherent mess. This is a common criticism of Nolan’s films in general, though not a popular one among his fans. It’s the first of a three-part series being hosted by Press Play about the editing of modern action sequences, and while this is the example of a “bad” action sequence, the next two will be examples of “good” action sequences. What struck me in particular about this essay was the comments that it engendered. Steven Santos brought up The Tree of Life, which I’ve argued previously has stylistic similarities in its editing to Michael Bay’s action technique. To wit:
Perhaps, because I spend so much time myself trying to figure out how one shot cuts into another, I see mistakes in pretty much every film. One of the aspects of these discussions on technique that bothers me, is that while films or filmmakers you don’t like are subjected to essays about their filmmaking abilities (Unlike others on this thread, I don’t object to the notion of you tackling this subject.), do you ever notice these problems in films you do like? I can name an obvious recent example for myself: “The Tree of Life”. A film which I thought ultimately was emotionally effective, but the more I think about it, the more I feel that film was a bit of a choppy mess in its editing that undercut how good that film could have been.
I would even acknowledge when great directors stumble during moments, as I mentioned with both Spielberg or Mann or, say, Scorsese in the opening and closing battles of “Gangs of New York” which demonstrate he is pretty incapable of staging of large-scale action sequences. Perhaps, what I’m getting at, is wondering why in the critical community filmmakers are criticized for their choices only when the film/director isn’t liked, but sometimes employ contradictory logic to justify the filmmaking of their own favored directors?
Again, the point is not to say that Bay and Nolan’s films are of equal artistic merit and ambition or even equally as effective. But Santos articulates with bingo-bango precision the cognitive dissonance I was trying to reconcile in my own previous post. As it happens, Santos also commented on a post over at The Man from Porlock, where Craig’s wonderful, succinct review of The Tree of Life (in which his concluding sentence pretty much sums up all that I thought worked well in the film) included a mention to my post:
The best action movies —The Road Warrior, Raiders of the Lost Ark, The Terminator, Casino Royale, et al. — show an intrinsic understanding of the holy trinity of space, time and movement. Despite having immeasurably bigger budgets and special-effects to play with (or possibly because of them), Michael Bay is oblivious or indifferent to these things. He’s forged a signature style of arrogant ineptitude that gives hackery a bad name. (The original hacks of Hollywood actually knew what they were doing.) Malick’s method is trickier to ascertain. Would The Tree of Life be more effective if it actually had standard scenes, rather than a series of fragments of scenes? Could the film’s central narrative stand without the avant-garde passages that surround it? That version of The Tree of Life might have made a good movie too. It wouldn’t have been a Terrence Malick movie, however.
That’s a fair point, and this is part of Santos’s response:
The main issue I had with “Tree” and “New World” is that while others see some innovative editing style, I see a director employing a team of editors (5 on “Tree of Life”) to try to shape something out of the material and not necessarily succeeding. It doesn’t come across as experimental as much as it seems chaotic. So when people talk about the film achieving grace, I think of how the editing and, for me, ham-fisted ending pretty much work against that feeling.
I think Craig’s description of space, time, and movement as a holy trinity is apt. That trinity seems to be at the core of what confuses Emerson about The Dark Knight (and by extension, a lot of other contemporary action films, I would guess). By the same token, however, why should action sequences (even those as fantastic as the ones in the Batman and Transformers movies) be bound by Promethean chains to concrete realities while impressionistic dramatists like Malick or Wong Kar-Wai get a free pass to violate space, time, and movement? Craig’s riposte — “the original hacks of Hollywood actually knew what they were doing” — gave me a chuckle, but between his asserting that “Malick’s method is trickier to ascertain” and Santos’s perception that a team of editors was trying “to shape something out of the material and not necessarily succeeding,” a few broader questions came to mind.
Filmmakers like Malick and Wong are more improvisational, shooting exponentially as much footage as they will actually use (giving rise to the hushed-voice reverence offered in anticipation of those six-hour director’s cuts and whatnot) and eventually finding the film in the editing room. In a word, their filmmaking is more or less intuitive. Which is fine. Great. If it works, it works. They are, for the most part, acknowledged masters who have earned the right to employ that methodology. The first big question that comes to my mind, then, is whether filmmakers like Bay and Nolan aren’t also intuitive.
A casual perusal of DVD supplements shows that most directors, when shooting a big budget action film, storyboard the sequences in advance. But even with that advance planning, the logistics of the location, post-production reshoots and inserts, and other little changes throughout the process often result in the final version of the film differing — often in significant ways — from the painstaking plans made during pre-production. But how much of pre-production planning is devoted to mapping out adherence to the 180 degree rule? How important was it to Nolan during the planning stage to have a two-shot of Harvey Dent and his SWAT escort, simply to lock in where they were sitting in relation to the people in the cab? Did he think of the action sequence in those terms at all? I honestly have no idea. I don’t know for certain, but I’d wager that Lee Smith had very little input during that stage of the production process. The editor of a film doesn’t necessarily have a significant role to play in framing, lighting, and the initial storyboarding of discrete shots and how they’re intended to be sequenced. Even the screenplay for The Dark Knight specifies certain sequential shot elements that made it to the final cut of the film; Smith isn’t credited as a screenwriter, so I think it’s safe to assume that when Smith sat down to cut the film together, he could only work with the material given to him (as opposed to having had a hand in shaping the raw material).
The problems that Emerson cites in his video essay seem to have less to do with the editing process itself than the planning process that went into the shots themselves. Nolan clearly had a fairly solid idea of the key elements of that chase scene right from the beginning, and during the shoot, things like composition and the way the shots would eventually relate to each other were probably tethered more to his vision as director and screenwriter (and perhaps Wally Pfister as cinematographer) than Lee Smith as editor. When Emerson takes the film to task for violating the 180 degree rule regarding where Harvey’s sitting relative to the shotgun SWAT, there’s not a whole lot that Smith could’ve done to rectify those problems. My guess is that the shots were the shots. That’s Nolan’s problem, not Smith’s. But is it really a problem?
Perhaps yes, perhaps no. As I said earlier, Nolan had a strong idea of certain elements that he wanted to play up in the sequence right from the planning stage. He may have even had certain framing and movement ideas in mind as well. I know that Nolan has a bit of a reputation for being meticulous — but meticulous about what? As Emerson demonstrates, he’s obviously not meticulous about clear causal relationships between shots in his action scenes. There is another approach to consider.
As Emerson postulates (and commenters elucidate), the shot sequence seems more like the panels of a comic book, in which the relationships between shots are created more actively by the viewer. Steven Boone comments:
[C]omic books are a static form that our mind’s eye animates and gives velocity. In a film, inelegant, abrupt cuts between two pieces of action are like speed bumps. That’s what makes TDK often feel like having a double-D tit plopped in your mouth for a hot second before your assailant slips it back in her dress, kidney punches you and goes speeding away on a motorcycle while flipping you the bird. Exciting in a certain kind of awful way.
Nolan knew what he was doing, just as the architects of the Iraq War knew what they were doing. Should we give Cheney and Rumsfeld an Oscar?
Leaving aside the issue of whether or not Christopher Nolan’s action direction is bad enough for him to be considered an international war criminal, Boone makes two important observations: 1.) Our brains animate and give velocity to the images in comics; 2.) Nolan knows what he’s doing. We already know that Boone’s assertion regarding how the human’s mind’s eye works is accurate; David Bordwell and Kristin Thompson have been writing persuasively about cognitive psychology’s relationship to cinema for a long time. As for the second assertion… well, that’s the Big Question, isn’t it? Does Christopher Nolan know what he’s doing? Is he aware of the intricacies of his technique? If he is, what can we gather from his deployment of said technique?
In a very practical sense, Nolan’s action scenes are an accumulation of moments rather than a contiguous, extended scene. The convoy departs from the station. The Joker shoots the cop. The garbage truck bashes the squad cars out of the way. The semi smashes the SWAT van into the river. The Joker attacks the convoy from the semi. The Batmobile smashes head on into the garbage truck. Etc. etc. etc. For the most part, the compositions are not beholden to the awful shaky cam that muddles so many contemporary action films. Most of the discrete shots in the car chase are actually rather static or fluid tracking shots. Emerson argues that in the details, Nolan botches his delivery, but Emerson does concede that the sequence conveys the gist of what’s important. It’s more of a gestalt approach to action filmmaking. The pieces aren’t apparently intended to fit together like a meticulously-crafted jigsaw puzzle that ultimately forms a completely clear, unambiguous portrait of everything that a viewer could possibly apprehend with his naked eye. It may be the case that this approach to action filmmaking is aesthetically invalid and boneheaded. At the same time, Emerson makes an epically sardonic pile of hay out of which direction Harvey Dent is facing when the van is rammed, as if it’s impossible to tell that the van he’s in just got hit, and that he is indeed being driven by those particular two cops that we saw near the beginning of the sequence. Only a true idiot would not be able to understand — in general — what is going on in this sequence. As such, I don’t think the sequence is truly confusing, as Emerson argues, even if it is logistically unsound and poorly crafted.
Filmmakers since the beginning of cinema have used editing to “paper over” the gaps between shots, using perceptual cues and manipulations to suggest continuity in the film where none actually occurs on the film set. In other words, for the first several decades of cinema (especially in the days before camera mobility), filmmakers depended upon the ability of the minds eye to animate and give velocity to visual narratives. Nolan certainly doesn’t seem to care much about traditional, historical framing and editing techniques, but he definitely has a sense for how to wring maximum impact out of those individual moments that the film’s fans remember. The semi coming out of nowhere to ram the van into the river got an audible gasp at the first showing of The Dark Knight I attended. Everyone certainly remembers the rig flipping over at the end of the chase. A lot of the discrete moments that Emerson deconstructs are, in my experience, actually rather memorable, even if they are not well-constructed in the traditional sense. What I want to know is if this scene — and these moments — would be demonstrably “better” if they were done more in alignment with the suggestions Emerson makes. The fact that so many people had no problem following what was going on in The Dark Knight (as opposed to so many of the Greengrass knockoffs, about which average viewers complain voluminously) seems to indicate two things. One, that viewers may be more lazy than they were in the past; they’re cinematically illiterate. Two, that Nolan’s construction of this action sequence is not actually all that confusion; that it draws its power from the impressionistic accumulation of potent, discrete moments. Both of these things can be true. Which is why it’s important to bring the discussion back to Malick.
Films like The Tree of Life and The New World are nothing if not oriented toward the accumulation of potent moments. Taken together, these moments are meant to accumulate an emotional heft with a metaphysical overtone… just as the moments of a Michael Bay or Christopher Nolan action sequence, taken together, are meant to accumulate a visceral heft with a realistic overtone.
Again, I should probably clarify that Terrence Malick is a bit more ambitious with the thematic underpinning of his technique, but that has as much to with his narrative and themes as much as the technique itself, which is undeniably the same as Bay’s and Nolan’s. But just as Bay’s application of this technique is taken for granted as risible hackery, Nolan’s is taken for granted in this video essay as lazy and cinematically illiterate — and it may actually be all those things. But if it works (and I don’t know how you quantify that in any objective sense beyond gauging how many people have a genuine problem with it), does that make it bad? Ineffective? For some viewers, like Emerson, undoubtedly. Should his perspective be given more weight because he has a more highly trained eye than that of the average moviegoer? Yes. Does that mean that his basic assertion — that the chase sequence is confusing — is correct? Not necessarily.
Malick is forthrightly trying to replicate the experience of memory in The Tree of Life, but in The New World, the story unfolds as it unfolds. When that film came out, writers like Nick Pinkerton eagerly unloaded his Sardonic Death Ray upon other critics who dared to question Malick’s editing technique:
And so Stephanie Zacharek, writing at Salon.com, offers the following chestnut to drag Malick down to the lowest possible aesthetic denominator: “[It’s] like a Tony Scott movie on quaaludes: Words and pictures are matched up in counterintuitive ways, and although the cutting is much slower than in Scott’s hyperactive showboating, it makes just about as much sense.” Even more baffling: Dave Kehr’s blog manages to tie together the “music video” cutting in The New World and Domino — I feel like I’m taking crazy pills! Zacharek can be a great read when she’s taking the time to find little niches of pleasure in unpretentious entertainments that aren’t quite swinging for the fence, and she’s a not-bad writer when her Pauline Kael impression is in check (most embarrassing is when, aping her idol’s slangy style, Zacharek borrows Pauline’s actual slang: “high muckity-muck” or ”bamboozled.” Is this 1935?). But her New World review gives full rein to her laziest tendencies; there’s just something about Texas Transcendentalist Malick that brings out the blithe, “over it” undergrad in his detractors.
Maybe I missed it, but nothing in that entire paragraph actually addresses Zacharek and Kehr’s assertions regarding The New World‘s editing style. I guess Pinkerton’s crazy pills eroded his ability to construct even the most basic counter-argument. Let’s also consider John Patterson’s obliquely titled piece, “The New World: a misunderstood masterpiece?” (As if he’s really not sure what the thesis of his argument is? I guess? Perhaps? Maybe? By chance?) Unlike Pinkerton, Patterson manages to devote a whole paragraph to the editing itself:
And then there is the editing. Malick extrudes his movies from the film-bins in the editing suite, “finding” as much of the movie there as he does on location. I’ve seen three separate edits (the 150 minute pre-release version that knocked me out, the 135 release cut (25 of those 30 viewings), and the Blu-Ray director’s cut of 172 minutes) and all strike distinct and equally wondrous variations on the same themes, yet seem radically different to one another at a gut level. This clearly suggests that Malick’s editing has nothing in common with the frame-fucking visual aesthetics of Tony Scott, as has been suggested by more than one fool. In fact, it has more in common with Godard’s jump-cuts, which once seemed so radical and disorienting but which have been absorbed and are now part of the common, comprehensible rubric of the form. Far from being meaningless or self-indulgent, there is insight, a mini-revelation, a deepening of meaning, or just a blessed surprise in almost every one of Malick’s cuts, which cleave in style to this rich filmic inheritance, whereas Scott is a creature of violent eye-ache, and little else.
With the caveat that Nolan doesn’t really seem to be all that interested in “frame-fucking” (although I’ll grant that “frame-fucking” seems to be Bay’s predominant cinematic interest) doesn’t the way Patterson draws a family tree from Breathless to both The New World and Domino also include, by extrapolation, the action aesthetics of Nolan and Bay?
I don’t know about “radical,” but “disorienting” certainly seems to be a relevant term to Emerson’s deconstruction of that chase scene. Did Godard deliberate set up each shot in advance just to milk the maximum, cohesive effect out of his editing in Breathless? No. Patently the opposite. But Godard is a genius and Bay is a hack, because when Godard edits the crap out of a scene in a low-rent genre picture, it’s genius and when Bay edits the crap out of a low-rent (if high price) genre picture, it’s hackery.
Oh, dear. It seems that I’ve gone circular.
In neither of those pieces do I see any mention of Malick trying to replicate the experience of memory or even human perception. As Craig quite correctly states, The Tree of Life could utilize conventional mise-en-scene and editing, but then it wouldn’t be a Terrence Malick picture. And I think Malick has a valuable perspective both in thematic consideration and in technique. But the problem is that if we’re willing to chalk Malick’s editing technique up to auteurist predilection and accept it as part of what makes his filmmaking so valuable… why aren’t we willing to do that with directors like Bay and Nolan? Craig explains it this way:
I realize that runs counter to how I usually roll — making fun of auteur worshippers — but so be it. Malick’s film has plenty of other qualities that normally make me retch: the aforementioned spiritual overtones, Freudian psychology, a mother-figure who is Pure and True, lots of oohing and ahhing over babies. The reason they didn’t, I think, is because I love to see an artist caught up in his own grand passions; oftentimes they catch me up in them too.
Another fair point, but one that raises more questions. If it’s valid simply to get caught up in an artist’s grand passions, can’t you leverage that explanation against virtually action film — films that are explicitly designed to catch the audience up in the audiovisual excitement, even if all that excitement runs counter to what the audience knows to be plausible or sensible?
The very point of Emerson’s video essay is that The Dark Knight‘s car chase is nonsensical, but what if most viewers get caught up in it anyway? What if all those details deriving from cinematic editing conventions literally don’t matter because Nolan is aiming — intuitively, if not intellectually and conceptually — for a singular effect that can be achieved without adhering to the 180 degree rule?
By the standards of the holy trinity of action filmmaking — space, time, and movement — Nolan and Bay are incredibly awful action filmmakers. As I’ve said in earlier posts, my own prejudice is toward films that follow the doctrines of that particular trinity. It’s one of the big reasons that Jackie Chan is my favorite actor; his best films are masterpieces of contiguous framing, editing, and movement within the frame. But even films like Project A, Miracles, Police Story, and Drunken Master paper over mild gaps for the sake of a gag or an accumulation of great moments. (I’m thinking particularly of the double instant replay of Chan sliding down the electrified pole in the mall at the end of Police Story.) I think it’s perfectly fair for someone to say that they personally loathe Michael Bay or Christopher Nolan because they simply hate the lack of attention those filmmakers pay to the traditional notions of action filmmaking. If I didn’t allow for that, I’d only indict my own hypocrisy regarding so many of the modern genre flicks I routinely denigrate.
That said, it would therefore become inconsistent for the same person to say he loves The New World and The Tree of Life without at least acknowledging the similarities in editing technique and how those things are used to similar effect. On a related note, I think it’s perfectly fair to debate whether or not the “accumulated moments” approach to action filmmaking is “better” or “worse” than, say, the action aesthetics of early Steven Spielberg or James Cameron. “Separate but equal” may be an egalitarian axiom in conducting film criticism, but it’s neither interesting nor insightful. I’m sure that when it comes down to it, what really matters is how effective each technique is on a case-by-case basis. But the questions of validity and merit as they relate to the raw techniques are always going to be relevant to those discussions. Emerson is not trying to play a zero-sum game with his analysis of The Dark Knight‘s chase sequence; as he says, he’s just trying to explain precisely why some people had a problem with it in quantitative terms. I respect the work. But there are a whole lot of assumptions and aesthetic biases that go into his analysis. For me, the great part of a video essay like this is that it should provoke questions about those assumptions and biases, not to mention that it ought to inspire people who enjoy cut-rate cutting to explain why and how it works for them. I’m still a bit on the fence about whether or not it does work for me. But goes equally for The Tree of Life and The New World as much as it does The Dark Knight and Transformers 3.