September 21st, 2011 / 1:30 pm

A Critique of Jim Emerson’s Recent “The Dark Knight” Critique

You’ve probably seen by now film critic Jim Emerson‘s critique of a certain action sequence in The Dark Knight (2008). Many, many people forwarded it to me; they probably also forwarded it to you.

Now, I have said many mean things about Mr. Nolan in the past. But I actually disagree to some extent with Mr. Emerson’s analysis. (Maybe I just disagree with everyone?)

So allow me, if you will, to defend my buddy Chris Nolan.

First things first. I don’t think that The Dark Knight is “a great film,” but I also don’t think that it’s a bad film, either. I think it’s “an all right film,” plenty watchable enough, cathartic and entertaining enough—and very little else. (I do take some issue with what I perceive to be a certain somber nihilism on Nolan’s behalf, but that’s another matter.) I’ve seen it a few times now, and even watched it again recently, along with Batman Begins (2005) (which I would argue is a bad film).

But let’s not argue about that now. (I will, if people want, write separate posts about Nolan’s Batman films, and rest assured that I will be writing something rather long and critical next year when The Dark Knight Rises assails us.) Let’s focus instead for the time being on Emerson’s critique, and why I think it’s…somewhat misguided.

So second first things first. I rather admire Jim Emerson as a critic—I have, for instance, fully enjoyed his Opening Shots Project, and anyone who names his blog after a Cronenberg film is fine by me. I also think that this new In the Cut Project, taken in the abstract, is pretty awesome (even if I keep mistaking it for an ongoing analysis of that weird Jane Campion movie). Watching a sequence slowly and analyzing it, in voiceover—and what’s more, doing it in order to teach others how to watch and analyze cinema—that’s totally, totally great. (I also like to think that it’s inspired by Mike Stoklasa.) Add to all of this the fact that Emerson name-checks William S. Burroughs, Stan Brakhage, and David Bordwell while doing it—terrific stuff.

Furthermore, I agree with many of Emerson’s criticisms of The Dark Knight. The first appearance of the Joker in that sequence, for instance, isn’t necessary, and serves only dissipate tension. Emerson’s critique is thorough and well done, easy to follow, and even occasionally funny.

So what’s my complaint?

It has to do with the assumptions underlying his analysis. Let’s look first at what Emerson is claiming. His thesis seems to me to be: Nolan and his editor (Lee Smith) aren’t using continuity editing, and that makes the action sequence incoherent.

The first point is incontrovertible—and Emerson amply demonstrates it—but I’m not sure that I agree with that second claim, so let’s start there. To be sure, the sequence doesn’t make sense according to the standard Hollywood realism. But is it incoherent? Impossible to follow? I hardly think so.

What happens in this sequence? The Joker ambushes the police convoy that’s transporting Harvey Dent. He does this by means of ramming things with a garbage truck and a semi truck, and by shooting various vehicles and people. Along the way, he takes out a bunch of police cars and a SWAT truck. Then Batman shows up in his Batmobile and the battle is joined.

I don’t think it’s that hard, really, to follow that line of action (broadly speaking). Even the particulars of the sequence are not all that confusing: the semi takes out the SWAT van. The Joker fires rockets at the convoy. Batman takes out the garbage truck. Of course, Emerson is exactly right when he says that the sequence plays hopscotch all over the axis of action, and he’s right when he points out how police cars appear and disappear, and he’s absolutely right when he says that it’s hard to tell which side of the police van Harvey Dent is sitting on (both sides, it would seem).

But a larger question looms: do those details really matter? In other words, if this sequence were shot and edited according to the principles of standard continuity editing, would the sequence be easier to follow? Yes, to some degree. But would that matter? Would that make The Dark Knight a better movie?

I’d argue: no. And correct me if I’m wrong, but I doubt that critics like Emerson would like it any better if it were. (I know that my complaints with the film transcend the coherence of its action sequences. Along the same lines, my criticisms of Inception weren’t rooted in Nolan’s failure to follow established shooting and editing conventions—far from it!)

What's more, Emerson stopped well before he got to the sequence's best moment: when Batman flips the semi over—which was, despite my immense dislike for Christopher Nolan's filmmaking, FUCKING AWESOME. Not as good as when the semi charges off the overpass in "Terminator 2," mind you—but still pretty fucking cool.

My second problem with Emerson’s argument is perhaps more serious. As noted, Emerson amply documents how Nolan and his friends did not use continuity editing; this point is indisputable. But that then begs the question: are they therefore using some alternative system? Emerson, despite the thoroughness of his analysis, never asks that question; rather, he applies the metric of continuity editing to the sequence and finds it lacking. This strikes me as somewhat unfair—reminiscent, to some degree, of dismissing an abstract painting for not representing.

Indeed, Emerson asks at one point, “Did they take all the footage and throw it up in the air?” Obviously, no. The sequence has its own logic, even if that logic is muddled—I mean, it does convey meaning, somehow. Emerson knows this, and even occasionally hints that he can see some alternative logic at work—for instance, when he points out the “twisting cheat” involving the semi truck as it bashes the SWAT van into the Chicago Gotham River.

Now, I’m not about to sit down and look for logic in Nolan’s sequence, because I like to think that I have better things to do with my time. But that question—what logic is at work here (if any)?—is entirely relevant to Emerson’s critique. Without getting too deep into it, it seems obvious to me that Nolan’s sequence is much less concerned with spatial integrity, and much more concerned with SHEER VISCERAL IMPACT.

Once more, for feeling.

Is this good? Bad? The first steps of some new way forward? The end of great cinema as we know it? Hell, I don’t know. But I do think that Nolan was, in this one film, in this one sequence, pretty successful at something—at least in terms of IMPACT. Credit is due where credit is due.

Meanwhile, a more important question is taking shape: despite Emerson’s critical assumptions, does it automatically follow that Nolan et al. should have used continuity editing? No, I do not think so.

For various reasons, over the past ten years (if not longer), Hollywood has been gradually moving away from the gospel of continuity editing. That obviously bothers many more traditional—one might say more conservative—film critics. And I don’t entirely disagree with their lament: on some level, Hollywood is abandoning a large portion of its artistic craft. (The best account I’ve seen of this so far is David Bordwell’s, in his 2006 book The Way Hollywood Tells It.)

But the current situation is what it is, and it leaves us with some interesting lines of thought:

  1. Why has Hollywood changed the way it works?
  2. What system or systems (if any) are its directors now using?
  3. What are the advantages and disadvantages of that new system or systems?

I would love to hear Christopher Nolan’s account of why he made The Dark Knight the way that he did, and what narrative conventions he followed, or thought he was following. And I would love to see a critic of Emerson’s caliber tackle these pressing issues, rather than simply chastise Nolan for not using continuity editing. Or, to put it another way, I’d be much happier with the argument (should someone frame it this way) that the system that Nolan did use (if any) is not as good as continuity editing, for reasons X, Y, and Z. But Emerson isn’t making that complete an argument.

Me, I don’t think it automatically follows that filmmakers should use continuity editing. It’s just one set of narrative conventions, with numerous advantages and disadvantages. Insisting that filmmakers always do things a certain way—whether explicitly or implicitly—limits film’s ultimate artistic potential. (Formal limits may facilitate the creation of great art, but they may also result in hackneyed, formulaic movies. Craft is a double-edged sword.)

As Emerson himself says in his video, “There are lots of ways to make a film, and lots of ways to make a mess.” I agree. But he hasn’t convinced me yet that Nolan has made a mess—or that the way he made his film is not, in fact, a way to make films.

(Of course on some level, you please must understand, I can’t believe that I’m writing all of this. Chalk one up for intellectual integrity? Or sheer orneriness…)

Tags: , , , , , , , ,


  1. bobby

      I wish Patrick Bateman was Batman instead. 

  2. Anonymous
  3. jim emerson

      Thanks for a reasoned rebuttal. I just posted a transcript of the video essay for easy reference to what I actually said and did not say:
      Nolan does, in fact, follow traditional Hollywood continuity much of the time, and that’s where it gets disorienting for me. Just when you think you’ve got something nailed down, he makes it ambiguous. I pose the same question you do: Do his decisions make the sequence more exciting or just more confusing?  I’d love to see someone make a case for the former, because I don’t see it. I’m arguing that some of the movie’s choices lessen its impact as action: “If the filmmakers are trying to involve us in a semi-realistic chase, why do they insist on distancing and disorienting us with these conspicuously nonsensical continuity techniques?” (I did like the flipping of the truck a lot — and, as I’ve said in its defense elsewhere, I don’t care that it required an explosive pump beneath the front end to make it work. It’s a very cool stunt. I stopped because I felt I’d made my points — and the thing was already 19 minutes long!) As I say near the conclusion of the essay: “Are we meant to look at this as abstract motion, like an IMAX Stan Brakhage movie? Or is it, as the filmmakers have said, more concerned with realism — photographing real objects, including actors and miniatures, in real space? We can see how it does what it does. The question is: What’s the result? How do these stylistic choices enhance or diminish the impact of the movie?”This was intended to start a discussion along those lines. I hope it continues…

  4. reynard

      the editing of that movie is some of the sloppiest i’ve seen, but i always assumed it was because they shot too much and were trying to fix fuck-ups, uninspired direction, a lack of planning, and a lack of coverage, plus, chris nolan sucks balls

  5. jim emerson

      Dang, I forgot to add that the creation of the essay was sparked by my reading this quotation from TDK editor Lee Smith, which begins the video: “”It’s quite easy to over-cut a sequence: make it visually exciting and lose track of what is happening and who the characters are… Where you can’t follow action, it’s not just action, it’s the whole movie you can’t follow. Action is very difficult, it has to be very carefully planned and conceived.”
      — Lee Smith, editor (“The Dark Knight,” “Inception”), interviewed in The Australian, October 30, 2010If that’s not an endorsement of continuity editing, I don’t know what is…

  6. nanlan
  7. Guest

      Recently another pop culture object was critiqued–by Wired–in terms of (basic) film theory: video games–specifically their cutscenes. I’m wondering how interchangeable film and cutscenes are from a theory perspective.

  8. deadgod

      [A]re they therefore using some alternative [editing] system?

      I thought Emerson’s thesis was that conventional editing was used for most of the movie and abandoned for the big-crash sequence.  –that, for the viewer, the movie was invested in conventional film-narrative coherence generally but in noisily gibbering spectacle intermittently.

      The question for Nolan from moviegoer Emerson wouldn’t be the naively tendentious ‘why are you communicating in this way?’, but rather, the more grounded but far more suspicious ‘what are the interests governing your code-switching in this way?’.

  9. deadgod

      Eh, in other words, I think Adam and Emerson are seeing the same jar in the big-crash, what, interruption, and disagree as to its effect/effectiveness.

  10. deadgod


      The question for at least some viewers wasn’t whether there was a pop in the guts – or at least a limp pimp-slap – , but rather, whether the “impact” was lasting and interesting or a tawdrily meretricious startlement.  For me, it was a boring movie, with laughably simplistic psychological force–and, when a punch or even a hand-clap sounds like a car accident, where is there to go when there’s a vehicular collision or incendiary explosion?

  11. deadgod


      invites; leads to; provokes; ushers forward; lays the groundwork for movement in the direction of contemplating a hazarding of putting forth a consideration of entertaining the possibility of asking

  12. Whatever

      there is something in here about if it is even worth saying, “this is how ___ failed to ___”.

      i think this movie was pretty tight fr a bloated hollywood corpse, as good as i could expect from the industrial context.

  13. Adam D Jameson

      I wish that, too. Let’s wish that together, friend!

  14. Adam D Jameson

      Thanks for replying! And thanks for that transcript.

      I haven’t looked yet at the rest of the movie to see to what extent Nolan uses continuity editing, and whether he abandons it simply in the action sequences. Which I didn’t understand to be your thesis, actually (if indeed it was—but that may also be my failure to understand).

      I pose the same question you do: Do his decisions make the sequence
      more exciting or just more confusing?  I’d love to see someone make a
      case for the former, because I don’t see it.

      I, too, would love to see it, but I don’t know if “confusing” is the right word here. I think the real question is: Is it better aesthetically? And I like to think that aesthetics can include confusion.

      Thanks again for starting this discussion; I’ve really been enjoying it. Your questions are all excellent ones! I totally agree with your (superb) implication, by the way, that Nolan’s action-sequence editing seems to contradict his emphasis elsewhere on plausibility (I myself wouldn’t call it “realism” or “semi-realism”; those terms are just too loaded/complicated for my sake). Indeed, I think that line of criticism merits much more development. (I’ll go so far as to repeat my fundamental criticism from my Inception post: Nolan adheres strongly to the dominant Hollywood conventions of our time, bringing little else to the table, and as such he does little to interest me artistically; I’ll argue that art requires both adherence to but also brilliant deviance from convention.)

      Many cheers,

  15. jim emerson

      Thanks again, Adam. I don’t think Nolan cares all that much about plausibility (I don’t either — something I get into in Part II), but as I detail in the video essay, I was quite honestly confused by the incongruities and ambiguities of this sequence, which seemed to me unnecessarily sloppy and vague. To paraphrase a Scanners commenter: I shouldn’t have doubts or find myself asking these kinds of basic spatial questions while watching an action sequence; I expect the filmmaker to be fluent enough to SHOW what’s happening. 

      I certainly don’t think everybody has to use continuity editing, but if you’re going to stage a chase like this and emphasize the importance of the relative positions of the vehicles throughout, and you’re using continuity editing to do it (as Nolan does — most of the time), then don’t go about it in a haphazard, half-assed way. If a filmmaker is setting up a photorealistic chase, shooting real objects in real space, then he/she would be wise to COMMIT to it and follow through. 

      Sure, I’m well aware that “most people” say they weren’t confused and found the sequence somewhere between “great” and “good enough.” My critical goal is to show why I found it needlessly muddled and inconsistent — not to conduct a poll of popular opinion. Obviously (I hope), I think “aesthetics can include confusion,” too — but as I say, how do the techniques of “The Dark Knight” — whether they’re meant to be confusing or not — enhance its effects? I’ve explained why I don’t believe they do. If someone wants to argue that the movie is demonstrably better without the in-depth compositions I find lacking (of the truck at the station, the interior of the van, the moments between the crash and the splashdown, the post-crash trajectory of the semi, etc.), then I’d love to hear their reasons. 

  16. deadgod

      [W]hile “The Dark Knight” plays by some continuity rules some of the time, it doesn’t do so consistently.

      not just dumb little continuity errors [. . .] What’s more troublesome […] are the violations of spatial integrity between elements within the frame or between shots.

      [T]his photo-realistic IMAX action picture plays fast and loos — sometimes — with certain narrative filmmaking techniques that help make action understandable.

      I thought that Emerson’s point was that the movie is mostly spatially coherent, in the sense that filmed objects are spatially consistent with respect to each other’s motion/stasis in some particular scene, but that, in (at least) the rocket-launcher sequence, that privileging of that kind of coherence is abandoned.  Rather than taking certain action sequences as having their own ‘aesthetic’ (without knowing what that ‘communication by virtue of form’ might be), Emerson understands those sequences to have been carelessly filmed/edited – rescuable only by ‘special pleas’ to some different ‘aesthetic’ – , and that, until this alternate ‘aesthetic’ is, not explained completely, but at least reasonably disclosed, he’ll understand these sequences to be accepted on the grounds of uncritically received sensation and not a cogency which, for him, is the connective tissue of successful film “action”.

      That’s putting Emerson’s response pretty strongly in my own words, but I think the paraphrase is accurate.

  17. Adam D Jameson

      I find this all very reasonable.

  18. Adam D Jameson

      Yeah, that quote makes no sense. The disconnect is just bizarre. I’d love to hear what they were really thinking…

  19. Adam D Jameson

      I want to believe.