Dunkirk’s central conceit — its most Nolany feature, if you buy my claim that Nolan’s films are intellectual exercises first and foremost — is its weaving together of three different stories told on different scales of time. There’s “The Air,” which follows a RAF pilot and takes place over the course of an hour, almost in real time. There’s “The Sea,” which follows one of the civilian ships that took part in the Dunkirk flotilla, and takes place over the course of a day. And finally, there’s “The Mole,” which takes place on the Dunkirk beach, and condenses a whole week.

If you were to re-edit Dunkirk into chronological order, you’d start with something like a quarter of the land story. Then you’d get to the last day of that week, at which point the sea story would start up, and you’d cut back and forth between them for another fifteen minutes or so. Finally you’d get to the last hour of that day, and the air story would start up, and then you’d cut back and forth between all of them until the end of the film (although the air story would end a bit earlier than the other two, and the land and sea stories eventually merge into a single thread). That’s not how Nolan edited it, though: instead, all three stories unspool more or less simultaneously. This means that as we cut back and forth between them, we’re leaping backwards and forwards through time.

One might fairly respond, “When, O Christopher Nolan, do you mean to cease abusing our patience? How long is that madness of yours still to mock us?” I guess you’re just toooo clever to tell us a straightforward story with one beginning, one middle and one end, huh? Are you too good for narrative causality, Christopher?

But Nolan has a way of making good on his precious filmmaking conceits. Interstellar notwithstanding, most of his high-concept ideas ARE good ones. And at their very best, they also make for electrifying viewing experiences. Rather than interfering with the audience’s emotional engagement, the philosophical trappings end up twisting the basic language of filmmaking in a way that makes everything more exciting, not less. In Memento, which is still probably Nolan’s best and most characteristic work, the reverse chronology is meant to illustrate a philosophical point on the impermanence of identity — but it twists the basic language of filmmaking so that each cut, and each new scene, places us in a threatening and unfamiliar place (because we never know how we got there). As a result the film achieves nail-biting levels of suspense. Some movies are gripping, some movies are cerebral. Memento is a very rare example of a movie that is gripping because it is cerebral.

And now with Dunkirk, Christopher Nolan has once again managed to hit us in both the cerebral cortex and the amygdala.

Heart:  Dunkirk, considered as an Adrenaline Pump (and one big awesome montage).

Let’s talk about the emotional side first. It’s instructive to think about another way that Nolan could have presented his story: what if they’d done it like Rashomon? That is, what if we saw “The Mole” in its entirety, and then “The Sea,” and then “The Air”? This would have made most of the same philosophical points, but you would have had a very different narrative arc. Rashomon is basically a series of short films. Each account has its own beginning, middle, and end, its own climax, and its own rising and falling action. And with the exception of the dead husband’s tale, they’re all narratively coherent: they would pretty much work on their own as short films if you wanted to screen them that way. You’re never left wondering who the characters are, or where they are, or why they’re doing what they’re doing.

This is not a backhanded way of accusing Dunkirk of being confusing! As it stands, the film makes total sense. But the unscrambled versions of “The Mole,” “The Sea,” and “The Air” would not have coherent narrative arcs. I don’t think they’d stand on their own as shorts! Maybe “The Sea” could work, but “The Mole” and “The Air” are too episodic and repetitive, and all three of the stories seem to have cut out too much down time. Mark pointed this out on our Dunkirk podcast: if you imagine “The Air” as a freestanding film, it would start with a bunch of scenes of the pilots playing shirtless volleyball (or cricket, as the case may be), so that we know who the characters are. But instead we’re dropped into the story in medias res, and after that it just jumps from dogfight to dogfight to dogfight. Watching all of that footage in a row, you’d be begging for some kind of transition shot — something to establish narrative and temporal distance between dogfight one and dogfight two. At the very least, you’d want Tom Hardy to pipe up with a line like “Another Messerschmidt? Come on, I like JUST shot down the last guy! This is ridiculous.” But of course in the actual film we don’t care about (or even notice) these missing interstitial pieces, because Nolan has carefully sculpted arcs that run across all three stories simultaneously.

These timeline-jumping arcs are not narrative arcs, mind you. They aren’t causally linked arcs of events. They’re affective arcs, rising and falling patterns of intensity, velocity, effort, and danger. One way to think about this is to consider the relative level of agency that all the characters have. The soldiers on the beach are basically passive victims: they can hardly do anything. (“The Mole” is honestly more of a disaster film than a war film, for this reason.) The sailors are doing something, but their action is pretty monotonic: they march steadfastly into danger, only occasionally reacting to some external event. (This is why “The Sea” is so much talkier than the other strands of the film.) And then “The Air” is basically a power fantasy video game: all pew-pew-pew all the time. Video games are very much about giving people a way to channel affective tension into heroic-seeming violence, right? I don’t mean the crude catharsis that people sometimes suggest, where you kill cops in Grand Theft Auto so that you don’t kill your shift manager in real life — just that if you’re stressed out after a long week, you can burn off a lot of that tension (at least temporarily) by stomping on Goombas. In Dunkirk, it’s often the case that emotional tension created during a slice of “The Mole” or “The Sea” will be released through an explosion of violence in “The Air.” Significantly, the three chapters do NOT have distinct musical identities. A big crescendo that starts in one timeline will usually end in one of the others.

Speaking of the music, I’d be remiss if I didn’t point out that, in at least a few cues, we actually hear the same musical material playing out on multiple time scales simultaneously. Listen to the ascending octatonic scales in “The Oil,” for example:  super slow in the bass, with a couple of interlocking faster versions in the middle and upper register… and then suddenly at the 5:00 mark everything synchronizes, and we get a cool Shepard Tone-y passage where the scale is played in all three registers in unison. I’d have to go see the movie again, but I’m guessing this falls at the point where the sailors are trying to outswim the burning oil, which is really the one sequence in the story that’s shared by all three of the narrative strands?

Lots of people have pointed out that, even for a modern blockbuster, Dunkirk seems to be an unusually tense film. Sustaining this level of tension in a normal narrative would be well-nigh impossible: how can you write a story that just gets tenser at every turn? You probably can do it (and feel free to nominate films that you think have pulled this off), but you run the risk of lapsing into ridiculousness. At some point, the audience will start saying “oh come on, ANOTHER twist? MORE tension?” One place that this always happens is when there’s ten seconds left on a ticking bomb, and about three minutes of screen time goes by before the hero manages to cut the wire. Done well, this is exciting… but even when it is done well, it’s kind of ridiculous (and even annoying). By slicing up the plotlines the way that he has here, Nolan neatly sidesteps these pitfalls. Yes, every time we cut back to one of the timelines, something tense is happening. But this isn’t because nothing else ever happens — it’s just that we’re checking in on these characters precisely when they’re up to something interesting. And temporal distortion is built into the structure of the film. When Tom Hardy shows up just in time to save Mark Rylance’s boat, this isn’t a ridiculous coincidence that the screenwriters ginned up to force an emotional response — rather, the moment of literal co-incidence, where the timelines intersect, feels like the fixed point around which narrative revolves. It’s not that “our” airman had to be wrangled into the right position to save “our” seamen, rather the reason we’re following this airman in particular is that he’s the one that saved this particular boat (and vice versa).

Now, take a moment to think over what we’ve said about Dunkirk so far. We have a bunch of little scenes, all chosen because they’re exciting. We’re sticking them together without regard for fixed narrative causality, but they still add up to some kind of a sense of a developing action. And we’re making liberal use of music to glue all of these scraps into a broader affective tapestry. Does this sound like any filmmaking technique that you’ve ever encountered before?

Maybe in the 80s?

But the difference between Dunkirk and a normal montage is that here, we do SORT of care where the little snippets come from. Montage fragments are totally unmoored in time (unless you actually stick, like, a clock or a page-a-day calendar into the shot to make sure that the audience knows when it’s supposed to be). The Dunkirk fragments are only relatively unmoored in time. The effect is to reinvigorate a technique gone stale — to remind us just how insanely avant-garde the filmmaking language of something like that Rocky sequence actually is. If I were Christopher Nolan, and an interviewer came at me with some variation on “so you like to mess with time in your films, huh?” I would grin my smuggest grin and say “Actually, ALL films mess with time. You just don’t usually notice it.”

Mind: Dunkirk, considered as an Epistemological Treatise.

So that’s how Dunkirk works as an art object, and why it’s so gripping to watch. But we’re only halfway through our exploration of the film. What about the high-concept philosophical conceit? Why tell the story on three time-scales at once? (This being Christopher Nolan, I can guarantee you that “because it’s gripping to watch” is NOT the only answer.)

I’m pretty sure that the film’s structure is meant to drive home the idea that “the story of Dunkirk” exists on a purely subjective level, just like our experience of time. I think that Pete had the right idea on the podcast when he pointed out that, rather than calling the film “The Miracle of Dunkirk” or “The Dunkirk Flotilla” or some such, Nolan simply called the film Dunkirk, leaving open the question of what actually was, and allowing each of the timelines to answer this question by telling the story of one particular Dunkirk. In the film, we’re presented with three different points of view on the battle, one lasting a week, one a day, and one an hour. Significantly, these aren’t keyed to particular characters. It’s not “3: Tom Hardy.” Rather, they’re keyed to environments. From the sea’s point of view, Dunkirk lasted a day. Even the world, then, experiences the battle on different temporal scales. There’s no such thing as a natural time frame that the battle “really” falls into.

We can go even further: the battle of Dunkirk didn’t really exist as a distinct event. Not naturally. There’s no non-arbitrary way to divide the retreat from Dunkirk (which is conventionally dated from 26 May to 4 June 1940) from the events in WWII that came immediately before and after it. Sure, you can say it started at the moment that the first evacuee boarded the first boat — but that doesn’t work, because after all both the man and the boat had to get there in order for that particular interaction of physical objects in the world could take place. So do we start when the order to evacuate was issued? When the DECISION to evacuate was made? When the Germans successfully flanked the Maginot line, making the evacuation inevitable?

And how would any of these boundaries be less arbitrary than the one drawn by Nolan’s “The Air” story, where the Battle of Dunkirk begins at the first moment that Tom Hardy encounters the Luftwaffe over the English channel, and ends the moment he sets fire to his downed plane on the Dunkirk beach? Like Pete said on the podcast, for that particular pilot, that WAS the battle. And although a historian seeking to explain the battle might have reasons for blending all of these experiences into a more detached God’s-Eye view (which would be something like Winston Churchill’s subjective experience of the battle, really), a filmmaker seeking to convey the battle would have reasons for focusing on individual experiences in all of their irreducible specificity, including the different timeframes.

Consciously or not, then, Nolan is using this film to revisit a philosophical chestnut about the way that ideas relate to objects. Daniel Smith explains it like this (in an attempt to summarize Gilles Deleuze, who in turn is maybe riffing off of Gottlob Frege):

I can attribute the proper name ‘Battle of Waterloo’ to a particular state of affairs, but the battle itself is an incorporeal event (or sense) with no other reality than that of the expression of my proposition; what we find in the state of affairs are bodies mixing with one another — spears stabbing flesh, bullets flying through the air, cannons firing, bodies being ripped apart — and the battle itself is the effect or the result of this intermingling of bodies. Sense thus has a complex status. On the one hand, it does not exist outside the proposition that expresses it, but it cannot be confused with the proposition… On the other hand, it is attributed to states of affairs or things, but it cannot be confused or identified with states of affairs, nor with a quality or relation of these states.

Which is to say that without some observer drawing a circle around a certain set of bodily collisions, and labeling it “Dunkirk,” Dunkirk has no sense — which is to say that there IS no Dunkirk, not in the way that we usually mean. If Nolan had presented all of the slices of the battle in chronological order, he would have subsumed these three different experiences of the battle into one master-narrative. “Here’s what really happened, and here’s how these different experiences of the battle fit into that overarching reality.” But Nolan refuses to do this. Instead, he points out the fundamental particularity of human experience. My Dunkirk doesn’t fit together with your Dunkirk. If it seems to fit into a broader aggregate, it’s only because we retrospectively smush all of our particular stories together into whatever kind of sloppy aggregate we can make from them. (It’s not a jigsaw puzzle, it’s a pile of branches.) And this kind of sloppy, ad-hoc smushing together is basically the experience that, as viewers of Dunkirk, we actually have. We’re putting the aggregate together as we watch it. “Ah, that boat in the corner of the pilot’s vision: that’s Mark Rylance’s boat!” But because we have to do the work of constructing it, we recognize that our version of the film’s “actual plot” is a construction. Perfect synoptic vision of the battle is not something that we can get. It’s not something that we can really even hope for.

In short, the film reminds us that — although real British soldiers were blown to bits on a real beach by real German bombs — “the Battle of Dunkirk” has only ever been an idea. And the same goes for “the Miracle of Dunkirk.” This explains the film’s curious closing pivot to media representations of the battle, where a (basically false) newspaper story of about a dead boy’s heroism is juxtaposed with Winston Churchill’s famous “fight them on the beaches” speech. Nolan seems to be telling us that these are both noble lies. They are ways of thinking about what happened, but in no sense are they “what actually happened.” But maybe that’s okay, because they have important pragmatic effects — the first by honoring a dead child’s ambitions and giving comfort to his family, and the second by revitalizing the British war effort. And anyway, there’s no such thing as “the real story.” A story, in so far as it is a story, is always distinct from reality. So maybe this story is good enough.

Coda:  Larghississimo, con dolcezza

I want to punctuate this piece with one more observation about the music. Most of the score sounds like a mid-90s Nine Inch Nails album with no vocals and slightly more expensive synths. Which is to say: awesome, but driving and abrasive and very obviously mechanical. There are just a couple of places, however, where film music’s typical big-R-Romantic orchestra comes in — first when the naval commander sees the flotilla of small boats (and says “Home”), and second over the heroic ending of Tom Hardy’s “Air” strand. Mark and Matt brought these passages up on the podcast as well, calling them “major chord” moments. Take a listen:

There’s more going on here than just the harmony! The melody here is a famous one: the “Nimrod” movement from Edward Elgar’s Enigma Variations.

But the already slow tune has been slowed down still further — with each pitch sustained for something like eight seconds. Played this slow, unless you already know the tune (and are sort of waiting for each pitch in the series to appear), you won’t even experience it as a melody. Indeed, Matt and Mark seem to have experienced the passage as a series of isolated pitches and intervals, beautiful in their color, sweet in their harmony, but semantically empty: bodies mixing with each other in the sonic state-of-affairs, rather than some kind of cohesive whole where every note is part of a bigger story. And the music that they heard is not redolent of the British Empire in the way that Elgar’s music (when you know that it’s Elgar), inevitably is.

What would it be like, I wonder, to go into Dunkirk knowing nothing about the “Miracle of Dunkirk” as that story is usually told? To hear the Churchill quote at the end without knowing that it was Churchill? Would we experience it as a story of national triumph? Or is this, like the Elgar melody, something that you need to know in advance, in order to see it at all?