By J. D. ConnorJune 1, 2019

LATER THIS MONTH, Avengers: Endgame will become the biggest movie of all time. It is also the culmination, so far, of the largest ostensibly continuous cinematic narrative. It is simultaneously, as Drew McWeeny explains, the fourth Iron Man, the fourth Captain America, the fourth Thor, and the fourth Avengers. Those unprecedented accumulations of money and story will be crucial to the launching later this year of Disney+, the company’s proprietary streaming service. At the same time, the management of those storyworlds and the audience expectations that come with them congeals in graphical emblems internal to the universe that can stand for the corporation at large.

Ultimately, Endgame will be the biggest movie ever in a certain limited way: reported global box office, unadjusted for inflation. It won’t be the biggest domestic movie (The Force Awakens’s $936 million seems safe). And Gone with the Wind is still out there holding the title for adjusted-for-inflation box office, although that position is dubious. But Endgame is, as of this writing, about $100 million away from Avatar’s $2.8 billion global gross, and it will surpass that even if Disney has to resort to the sorts of shenanigans it did to get A Wrinkle in Time over the $100 million mark. The headline number is large, and Endgame may even end up becoming the first “$3 billion movie.” Where that money goes is not exactly a mystery, but it is moderately confidential. To make and market the film, Disney likely spent more than $500 million. Another chunk went to the theaters, a lot to Robert Downey Jr., and there were cast bonuses when the thing hit $1.5 billion. The Russo Brothers, who directed, likely have decent participation in the backend. But most of it will go back to Disney.

There are, for now, 39 movies that have made over a billion dollars at the box office. This is a group that tilts overwhelmingly to recent films, where that money is made largely abroad. The oldest is Jurassic Park (1993); there have been 11 in the past three years. Only The Dark Knight (2008), Rogue One (2016), and Black Panther (2018) made most of their money domestically, and all of them just barely; The Fate of the Furious (2017) made more than 80 percent of its $1.2 billion in the rest of the world. (Domestic includes Canada, fwiw.) No other national cinematic industry is in the blockbuster business in the way that Hollywood is — none of these movies originated outside of the US studio system, although China has certainly come close to producing an entrant into the billion-dollar club. Wolf Warrior 2 (2017) made $870 million, just about all of that locally; The Wandering Earth (2018) made almost $6 million of its $700 million in US theaters before moving to Netflix.

None of this numberwanging includes revenue from TV or DVD or VOD or toys or T-shirts or theme-park rides or marginal propensities to subscribe to proprietary streaming services or imputable increases in the consumption of previous titles in the same IP universe as you try to remember just what the fuck happened in Infinity War. If you think about it for a little while, you realize that different movies activate different parts of the revenue stream differently, and that this is why there is a Cars 3. And a Planes. And a Planes: Fire & Rescue. Endgame may end up the largest single movie by some metrics, not by others, but one reason it has leapt to the fore of discussions of popular monoculture is that by any metric, the Marvel Cinematic Universe (MCU) is the biggest, well, thing in movies: 22 movies tallying $21 billion in box office so far, a decade-spanning, uninterrupted string of commercial hits.

And yet despite all that, the part of Disney that makes movies (Studio Entertainment) accounts for only one-sixth of the company’s revenue and just about the same percentage of its profits. Of course, the parks and the toys and the shirts and the television networks feed off the movies in important ways, but it’s not a foregone conclusion that within Disney, moviemaking will be more important than Imagineering or consumer goods design. Movies are necessary for Disney but they are not, in a strictly financial way, dominant. Their corporate-existential importance emerges from a struggle to articulate corporate goals, internally and externally. Currently, the central corporate goal at Disney is the proper articulation of audiences in time and space, to allocate and monetize bodily presence or temporal and spatial distance. No other aspect of Disney’s corporate portfolio can yet make the case for being the site of that articulation, and so for now, the MCU lies at the heart of Disney, and the stories it tells are in part a complex web of stories Disney is telling itself.


One of the central occasions for that sort of corporate self-narrating is the commemorative event. In the summer of 2018, to celebrate Marvel Studios’s 10th anniversary, USC hosted a series of screenings and conversations. Four movies — Iron Man, The Avengers, Guardians of the Galaxy, and Black Panther — each with the director in attendance (Jon Favreau, Joss Whedon, James Gunn, and Ryan Coogler), usually with other key department heads, and always with producer Kevin Feige. It was Hitchcock/Truffaut for the superconglomerate era. [1]

The MCU project that emerges from those extended discussions has four tiers: franchise foundation (Iron Man), expansive fusion (Avengers), characterological confidence (Guardians), and representational certification (Black Panther). Each was crucial, as much for its process as its product. The foundational process is a weekly meeting of all department heads to discuss design. This creates a shared visual language and allows for unprecedented cohesion throughout the extended production process. Without that cohesion, it would be difficult to manage such gargantuan projects, but it would be impossible to manage the intricate internal differentiation among several of them simultaneously while they are at various stages of development.

Expansion established a process by which creative workers could make demands that were small and personal, even within enormous projects. Such demands could always take the form of Easter eggs, but because they would emerge from intense collective discussions they took the form within the production discourse of logical extensions. The five-minute discussion of the shooting of the shawarma tag at the end of Avengers is a superb example of how Feige’s total commitment to his director’s tonal and narrative desires — desires that are at the same time reflections of the production process, as Whedon admits — can activate a chain of other creative workers (actors, editors, costumers, et cetera), even after the movie is supposedly locked.

The success of that mutual commitment underwrites the series’s growing confidence. The crucial launch here was Guardians, which was a property no one had heard of. The Marvel back catalog has thousands of characters, an infinitude of relationships. Some characters are dumb, some are great, some are utterly minor, some are well known. Launching Guardians was essentially a test: if we can do this, we can do anything. It made $773 million on a budget of $170 million. Only Iron Man 3 and The Avengers had earned more at that point.

Marvel’s “Uberframework” is an attempt to capture every narrative relationship in its vast library of characters. Each movie — even “Endgame” — only activates and reconfigures a subset of those relationships. Somewhere at the heart of this graph lies Wolverine.

That confidence amounts to an uncanny ability to pick one’s spots: to know, more or less, that this story, with this talent, at this budget, is likely to make this much, and so it should be placed here. That sounds vague: ask yourself how much you should laugh at the very idea of Ant-Man, at who other than Paul Rudd could pull it off, at how expensive it should be, and at when you should release it (high summer, when the audience is primed for goofiness and not yet thinking about going back to school).

Twenty-three movies in — including Spider-Man: Far From Home — and that confidence extends to an ability to push back at the Disney machine. When Guardians director James Gunn was fired by Disney for some exceptionally gross Twitter jokes, he was quickly hired by Warner Bros. to rescue the DC Extended Universe by essentially rebooting Suicide Squad. A year later, and Disney studio head Alan Horn hired Gunn back.

Still, Feige has been utterly judicious about when and how to push. Over the years, fans (and others) have pushed for a less white, less male MCU, and Feige (and others) have managed to create an underdiscourse, in which the limits of the MCU’s representational efforts stem not from his convictions but rather from constraints placed on his own fandom by longtime Marvel head Ike Perlmutter and conservative forces on what was called the “Marvel Creative Committee.” Feige was able to get Perlmutter and the committee out of his way in 2015, and the next four films out of the pipeline would be developed, written, shot, and edited without their input. It’s no surprise that those four films happen to be the “boldest Marvel has ever made”: Guardians 2, Spider-Man: Homecoming, Thor: Ragnarok, and Black Panther.

Here the crucial installment is Black Panther, which seemed to prove that the whole machine could just as easily work based on African diaspora superheroes, with departments largely headed by women of color. Black Panther offers a vision of merit deferred. In place of lamentations about the empty pipeline, here was a movie that suggested, convincingly, that the representational revolution was at hand and only required Hollywood certification. The industry was clearly ready to endorse that vision of incremental revolution, giving Oscars to both Ruth E. Carter (Costume) and Hannah Beachler (Production Design). Those two, along with an award for Black Panther’s score, were the MCU’s first wins.

This story — from foundation and expansion to confidence and representation — has been emerging within the MCU. At the end of Endgame, Tony Stark is dead, Steve Rogers is old, and Thor has a new home among the more ridiculous and sentimental Guardians of the Galaxy. Replacing the foundational three white dudes are Captain Marvel, a new Captain America, and Black Panther.

That agenda is encapsulated in the movie’s two frame stories. Endgame begins with Hawkeye’s family having a picnic, with him doing paternal things like teaching his daughter to shoot a bow and ordering toppings for his hot dogs. They all vanish, powderized like so many from the end of Infinity War, when Thanos exterminates half of all life in the universe. Hawkeye has been unlucky: half of all life includes his wife and three kids. The movie ends with Captain America dancing with Peggy Carter in a post–World War II period living room to an instrumental version of “It’s Been a Long, Long Time.” Together these scenes constitute an external frame in which the destruction of the straight, white, American nuclear family finds its compensation in the establishment of the straight, white, American couple. The movie is thus structured as a comedy of remarriage, that classical Hollywood genre Stanley Cavell detailed and that included Bringing Up Baby, The Philadelphia Story, and His Girl Friday. One of that genre’s principles was a distrust of reproduction — kids get in the way — which Endgame can’t quite make itself believe.

So inside that outer frame is a second one, beginning with Tony Stark believing he is going die only to be spared, and ending with his funeral and his final words to his daughter. He has made what the first Avengers movie calls “the sacrifice play,” snapping Thanos’s forces out of existence at the expense of his own life. The movie is thus structured as an envoi, finally delivered. The irony here is that Steve Rogers hasn’t sacrificed but has availed himself of a Starkean loophole in order to be with the woman he loves. Their stories are complementary, and closed.


If this what the MCU has been, what sort of model — for narrative, for business — does it offer? That is, how might it understand itself or be understood within the current mediascape? As individual entries in the superhero genre, the model is that of the great producer-units of the Hollywood past — the Freed Unit behind the MGM musicals, the Val Lewton horror films at RKO. As a whole, the MCU model is closer to that of serial television drama, with a strong executive producer and a shifting staff of writers and directors. But neither of these is quite right. The “superhero” is probably something beyond genre now — more like a mode, the way animation and documentary are. Within that mode, there are heist movies, political thrillers, family melodramas, and the usual origin stories and guilt-trips. And if the MCU is megatelevision, it is megatelevision in a slightly old-fashioned version, more typical of the broadcast-and-cable era.

If the MCU is old-fashioned megatelevision in its production, it is also a victim of old-fashioned television its storytelling. The “time heist” that makes up Endgame’s Act II is the cinematic equivalent of an end-of-season clip show, with various heroic subsets dropping in on their earlier selves, witnessing, retconning, and fanning over their earlier triumphs. “Remember when Quill stole the power stone? What a dork.” “Remember when the Aether ended up inside Natalie Portman? You probably don’t. Anyway, here is a shot of Natalie Portman which sure looks like unused footage from Thor: The Dark World. How Rocket gets the Aether out of her will be handled offscreen and never spoken of again.”

This immersed retrospection suffuses Endgame from its frame stories through its midpoint pivot when Bruce Banner promises Tilda Swinton that they will return the Infinity Stones to their original timelines. Narratively, Endgame amounts to an attempt to activate as much of the back catalog of the MCU as possible, without requiring audiences to have those moments at their fingertips. Or, to nest this problem in the representational system of these late Avengers films, to have them attached to their intellectual Infinity Gauntlets.

That problem — knowing what to refer to, what to activate, what to assume or imply or alter or efface — is central to any late entry in a continuing story tradition, but even more so when that tradition has been corporatized and managed as an extensible universe of intellectual property. It is only compounded by the continual addition of characters. One of the great things about killing off half the universe in Infinity War is that you don’t have to manage that overburdened character roster until everyone returns in Act III of Endgame. This is a kind of narrative abeyance that becomes increasingly necessary as franchises get bigger and add more and more team members (and possible villains). Fast and Furious handled this in a couple of ways — by liquidating nearly the whole cast for Tokyo Drift, by having half its heroes turn heel, by (metaphorically) killing off its co-protagonist. But eventually, narrative Malthusianism caught up and they resorted to sticking people on islands or in the hospital for long stretches.

If it would be impossible to narrate all these characters, it might be possible to display them in a kind of shorthand image, an assemblage. Outside the United States, The Avengers was known as Avengers Assemble, which was better and more accurate. The whole of the film was about getting everyone together, a process that culminated in a couple of showy shots — one that spun around the group (Whedon called it the “dolly shot”) and one that zipped from superhero to superhero (the “tie-in shot”) — in addition to the bathetic shawarma shot that showed them exhausted. Throughout the movie, that narrative process was shadowed by its allegory, Agent Coulson’s collection of near-mint vintage Captain America trading cards. (I discuss these in Hollywood Math and Aftermath.)

Such emblems of assemblage are especially endemic to the Avengers. (Buying a lot of vowels there.) They include the slo-mo line of battle and the end-credits statue from Age of Ultron, as well as the mega-clash from Captain America: Civil War. Such summary images don’t really align with the narrative engines of Disney's Star Wars universe, which are, at the macro scale, about inheritance, desert, connection, and imitation. (Social totalities in Star Wars end up rehearsing Triumph of the Will.) Nor do these emblems fit the central lines of the Pixar story, which turns around obligation and betrayal, usefulness and play, demographics and typology.

The last two Avengers movies have centered on the assemblage of the Infinity Gauntlet, which Thanos will use to destroy half of all life in the universe. Over at the media site REDEF, Özgür Gültekin produced some very enjoyable key art for an essay on Disney+ featuring Robert Iger morphed into Thanos. The five visible stones are labeled Hulu, Pixar, LucasFilm, 20th Century Fox, and Marvel. The image is untitled, but the html lets us know that internally it’s called Bobnos.

One can easily imagine that picture making the rounds inside the company, not simply because lots of people hate the boss who snapped his fingers and vaporized half of Fox’s film slate and thousands of their fellow employees, but because it captures one of the ways in which creative discourses grapple with strategic questions.

At the climax of “Endgame” every living hero in the MCU streams onto the field of battle through Doctor Strange’s “sling rings,” a figure for what Disney+ will offer for a monthly fee.

The vision of infinite Bobnosian corporate power has its inverse analogue in Endgame’s grand climax, when more than a dozen Doctor Strangey “sling rings” open, and all of the resurrected heroes from around the galaxy start pouring onto the field of battle. This is an emblem of assemblage no longer motivated by nostalgia (the Captain America cards) or the classicism of the newest Laocoön at the end of Age of Ultron, but by total spatiotemporal openness, a utopia of options. Where to look? Over there! It’s Howard the Duck! Over there, the Ravagers! Those portals are not mere channels that offer up the resurrected heroes that fans have been clamoring for. They are glimpses into rendered worlds to which those heroes might return for more stories, more sidekicks, or more-ness as such. At the same time, they figure audiences, assembled according to the rhythms of this meta-blockbuster: an audience absent, an audience present, here and now. These portals can dump superheroes and superaudiences onto this screen and onto any screen, anywhere anytime. Disney+ is coming, whether it is snapped into existence by Bobnos or demanded by fans or both. This is what it will feel like.


But in this gathering-of-the-juggernauts moment, which and what and who will matter is as opaque as the portals to those other worlds are transparent. Comic storytelling has been dealing with this problem for years. From “who would win in a fight?” what-ifs to memes like the one that had Ant-Man going up Thanos’s butt then enlarging and killing him. At this stage in the MCU, the universe is populated not by the limited-power specialists of X-Men, but increasingly by heroes who find new routes to omnipotence.

The rules of their interaction are tacit and can feel malleable or jettisoned. Surely there is a story bible somewhere that lays it out, but just as surely, it’s hard to know why anything is going to happen next. The result is a cinema of moments — what animators call “moving holds” — in which cameras are futzing around for no good reason and folks are just-mobile-enough not to be statues.

There are different ways of reacting to that. At the most credulous end we have Ryan Arey going through “209 Easter Eggs” in Endgame for ScreenCrush. Most of the eggs are merely references, but the affect is more important than the terminology. “The last fight is like all of your toys and all of your friends’ toys together for one big battle. There’s so many great moments in this fight […] In this battle, everybody gets a hero moment.” But if one doesn’t feel that the battle is “40 minutes of glorious, fan-service fun,” those moments can seem like unmotivated pandering. As Esther Rosenfield puts it:

The movie takes it for granted that you're here for the moments, and to hell with how they're meant to fit together […] You're meant to appreciate each one and then instantly wipe it from your memory. If you hold any two scenes in this movie in your mind simultaneously, the whole thing falls apart. It’s a film that resists being viewed. It’s un-cinema.

There is every reason to hate what these movies have become: notes toward a supreme fan service. For those inoculated to the charms of the comicbook superhero movie, Endgame seems to sync entirely too well with the conquest of cinema by (mere) content. Rosenfield again stakes out the sharpest version — “It's not a movie at all. It's just Content. And now that Disney's one big step closer to swallowing an entire artistic medium, steel yourself for a hell of a lot more of it” — but even Matt Zoller Seitz wants to regard Endgame as such a landmark. It is a “seismic cultural event,” “the decisive defeat of ‘cinema’ by ‘content.’” “It's hard to imagine subsequent MCU features, even gigantic ones, exceeding Endgame in impact, because no matter how huge and shiny and hyped they are, they won't be first, and they won't feel new.

This is the discourse of novelty exhaustion, and it is almost always wrong. Something will be bigger than Endgame. Probably incrementally bigger not quantum-leappy bigger, but so what? No one imagines that Gone with the Wind was the last movie of the Studio era because it was, and still sort of is, the biggest studio movie ever. More to the point, even the mode that Marvel has been working out is far from exhausted. Spider-Man: Into the Spider-Verse (which, I know, wasn’t technically MCU) was utterly surprising technically; Ragnarok was tonally unanticipated; Black Panther’s Killmonger was too morally compelling to be an instance of playing it safe. And they haven’t yet made a full-on musical. Anyway, talking rock monsters fomenting revolution in New Zealand accents hardly seems like the eradication of cinema for content.

Lamentations about “content” are not a matter of historical actuality but of discursive limitation. They appear when “real” creative work seems to be undervalued and when someone reaches the ends of their ability to regard practicality as anything other than degradation. But even then there are two flavors of attack. On the one hand, calling this “content” can be a way of upbraiding the willingness to settle for the disposable, the interchangeable, the mercenary, the pandering, and so on. (This is more Rosenfield’s line.) On the other hand, it can also mean a sort of false consciousness in which, despite whatever its creators might believe, they are in practice simply feeding the maw of distribution. (This is more Seitz’s line.) Then the claim is no longer that these particular instances aren’t good or to the critic’s liking, but that the system is set up to corrupt the significance of its objects in such a way that no broadly available reading can meaningfully redeem them.

The critical mistake here is to believe that the folks who have been making these movies don’t know that this sort of reckoning is upon them. As utopian as that moment with the sling rings might seem, it posed as many problems as it solved. If the heroes are no longer rankable, then the “hero moments” are no longer necessary to a rising course of action. This narrative weakness is both the practical problem of the third act and, inevitably, the ultimate threat to the MCU’s rhythmic cycle of releases. It may seem strange to thematize your own creeping dissociation, but that is what is going on in Endgame: You think this doesn’t make any sense? Have you considered … time-travel, a narrative device that is destined to create paradoxes, loops, and logical inconsistencies? The movie can’t stop bullrushing past its own confessions that it has no idea how time-travel might work, from jokey-jokes about other movies about time-travel (although neither Primer nor Déjà Vu) to the “aha moment” when, in his off hours, Tony Stark “solves” the problem by telling the computer to take a Möbius strip and “invert” it. Whatever, dude.

Time-travel may be conceptually intractable, so the MCU works through it practically, generating and dissolving its own local timeloops. By slating Spider-Man: Far From Home so soon after Endgame, it was obvious to everyone with a release calendar that Peter Parker was going to come back. But even knowing that, Marvel held off releasing the version of the Spider-Man trailer that really spoiled Endgame for a couple weeks. Even then they ran it as a proper trailer, after the movie. Some of Brie Larson’s action in Endgame was shot before her “debut” in Captain Marvel. And, again, knowing that schedule only enticed further speculation from fans who wondered how her unique powers might turn the tide against Thanos.

That practicality may buy the folks involved enough time to sort out the consequences of their own actions. Will they have been mere enablers of Bobnos and his avatars? Will the utopian emblems of association have been mere distractions from the real extraction at the heart of this system? Does the cult of industrial competence — the respect paid to the fact that this job has never been done this well for this long ever — override the question of whether it has been done well enough? Will its moments have been catalyst enough? Enough for what? To save the world, right? Is that asking too much?


J. D. Connor is an associate professor of Cinema and Media Studies at the University of Southern California. He is the author of Hollywood Math and Aftermath:The Economic Image and the Digital Recession (2018) and The Studios After the Studios (2015).


[1] Working critics don’t read academics with any seriousness — most glaringly no one seems to have cited Martin Flanagan, Mike McKenny, and Andy Livingstone’s The Marvel Studios Phenomenon or Derek Johnson’s work on Media Franchising — but the least they could do is take seriously the things that creators have to say when they are being reflective. Most don’t even do that.

LARB Contributor

J. D. Connor is an associate professor of cinema and media studies at the University of Southern California. He is the author of Hollywood Math and Aftermath: The Economic Image and the Digital Recession (2018) and The Studios After the Studios (2015).  Beginning in April 2023, J. D. Connor has written the City of Industry column for LARB: City of Industry offers monthly dispatches on the collisions between the art … the industry … and the experience of contemporary Hollywood. The art is often underestimated and demands closer analysis. The industry is undergoing profound and rapid changes that demand explanation. And our experiences of movies and TV and everything that has grown up around them demand more precise description. City of Industry tries to do all those things at once. Neither an insider’s guide nor a view from nowhere, this column is about what it means to think inside the Hollywood slipstream.


LARB Staff Recommendations

Did you know LARB is a reader-supported nonprofit?

LARB publishes daily without a paywall as part of our mission to make rigorous, incisive, and engaging writing on every aspect of literature, culture, and the arts freely accessible to the public. Help us continue this work with your tax-deductible donation today!