Kevin Nguyen | The Verge

The world’s oldest art, now in 6K IMAX

2026-04-14T15:46:05-04:00

The earliest paintings were made over 32,000 years ago — the very first forms of art and culture. They weren’t discovered until 1994, when cave explorers in France stumbled into the Chauvet Cave. More than a decade later, filmmaker Werner Herzog was allowed rare access to the highly guarded prehistoric site to shoot what would become the lauded 3D documentary Cave of Forgotten Dreams. It is a strange and moving film, one wherein Herzog convincingly argues, in his heavily enunciated German accent, that these caverns are the birthplace of “ze mo-dern hu-man soul.” Fifteen years since its premiere, the movie has achieved a cult-like status, and for a short time, you can now see it back in theaters as a 6K restoration on IMAX screens — housed in some of the largest and loudest cineplexes in the world.

When I first saw Forgotten Dreams, it was at a small independent theater in Seattle. In 3D, the experience was intimate — limestone stalactites and stalagmites press toward your face — and appropriately claustrophobic. Rewatching it at a press screening at the AMC Lincoln Square (the only “real” IMAX theater in New York City), the effect was, frankly, overwhelming. The clarity and detail of each grain of limestone wall, suddenly maximized across a screen that the human eye can barely take in all at once, makes Chauvet feel even more alien. The walls almost resemble skin — freckled with crystals, scarred by time.

“This is what 3D was made for,” raved one critic. Yet, Herzog didn’t set out to make a 3D film. In fact, he doesn’t really like them. Even seeing James Cameron’s Avatar, lauded as the hallmark 3D film of this century, Herzog was unimpressed. (“Avatar could be in 2D in a big theater,” he tells The Verge.) But when he was in preproduction, Herzog was allowed a visit to the Chauvet caves two months before filming and was struck by the experience of seeing the cave paintings up close. “All of a sudden I discover there are wild bulges and recesses and caverns and rock pendants — a world that is only existing in 3D because the painters 32,000 years ago utilized the formations,” he says. The cave painters — arguably humankind’s earliest artists — did not work on flat surfaces; in fact, the shape and texture of their canvas informed how and what they painted. “A bulging rock is now a bulging neck of a bison that attacks you,” Herzog says, as an example. Herzog may not be a fan of 3D films, but it suddenly made sense for his cave painting doc.

Still, shooting in 3D in such a specific environment came with its own challenges. For one, no 3D cameras existed that were small enough to be brought into the Chauvet cave, so they had to be created. “[The film] was shot in 3D with our own camera, our own data management, our own ‘brain,’” Herzog says, crediting Estonian filmmaker Kaspar Kallas for building the equipment (“a very, very intense and wonderful man”). But the setups, though custom, were also held together by glue and gaffer tape. (If you want more nitty-gritty details, there’s a great technical writeup by the film’s director of photography, Peter Zeitlinger, available at Mubi Notebook.) The movie was shot in 2K with SI-2K cameras, GoPros, and amateur-grade Canons. But today, the standard is 4K, and if you’re putting things on a screen as big as IMAX, you can go as high as a resolution of 6K or even 8K.

(Another fun fact: One of Sam Mendes’ James Bond movies, Skyfall, is often credited as being the first feature film to use drone footage — something that you see ubiquitously today. But Cave of Forgotten Dreams, released a year earlier, is actually the first. The crew hand-built a camera rig that could attach to a drone.)

In 2010, James Stewart, a 3D producer, was brought on to help refine the original film’s 3D experience before an early version had screened at Toronto International Film Festival. A decade later, he began overseeing the team that would restore it for IMAX, a process that began during the covid-19 pandemic and would stretch out for another five years. But Stewart’s enthusiasm for the film has never waned; in fact, this newer, more immersive theater experience has him even more excited. “In IMAX, it’s just mind blowing,” he says. He estimates he’s seen the movie over a hundred times, and even still, each viewing stuns him so much with its clarity that it makes him want to “lick the cave walls.”

“Cave of Forgotten Dreams, you can show it 150 years from now and it still will be completely fresh”

The work of film restoration is not unlike the work of archeology depicted in the film: the act of preservation. Stewart led a small team of under 10 people working over five years. From the extracted raw 2K footage, the movie needed to be rebuilt frame by frame, and Forgotten Dreams, being a 3D movie, was actually double the work, since there is a separate stream for the left eye and the right eye. (Aside from the meticulous patience required, the team also needed new software to extricate the outdated and oddly specific codecs of the original footage.) Finally, there is the effort to get that footage from 2K to 6K, which took experimental software and a lot of different hardware “to scale it up without just blowing it up,” Stewart says, clarifying that no AI was used. The team also rebuilt the film’s audio, a similarly painstaking process going from a 5.1 mix (six speakers) to a Dolby Atmos mix (up to 100 speakers).

According to Stewart, the first Avatar kicked off a “3D revolution,” though in the years since, it’s seemed like filmgoers’ interest in 3D has waned. That interest didn’t translate into 3D TVs for home cinema, and even each successive Avatar film, though still billion-dollar behemoths, has been less and less lucrative. From the way Stewart talks about “the greater 3D community” who work on these projects, it sounds like a small group of people with specialized skills. He’s also critical of the ways 3D is used in films — though he doesn’t name the bad examples. Still, he believes that for great filmmakers, it can become a powerful tool for storytelling. He cites Martin Scorsese’s Hugo, Ang Lee’s The Life of Pi, and a pair of Wim Wenders documentaries, Pina and Anselm, as “masterful” examples of 3D done right.

Stewart is proud that Cave of Forgotten Dreams regularly appears as the runner-up on lists of the best 3D films of all time. (Behind Avatar, of course.) But he still thinks the strength of the movie is Herzog himself — “the way he speaks and his writing and the way he tells the story.” And as awe-inspiring as the caves are, it’s just as impressive that Herzog and his small crew were able to capture so much given the limitations. The caves were dark; their equipment was limited; for preservation reasons, they could only shoot for a few hours a day over the course of a week. “It’s like, ‘Go make a film and then by the way, it has to be a transcendent visual experience.’ They’re like, ‘Well, you know, we’re lucky to get any footage at all.’”

Herzog himself was not hands-on with the restoration. (He admits “the digital, real work, I do not understand.”) But seeing his film for the first time in IMAX was “a very deep experience.”

I asked Herzog if he often revisits his films — there are dozens of them, many considered classics. Outside of attending retrospectives, he rarely rewatches his work. When he does, though, he is happy with how they’ve held up. “My films do not seem to age,” Herzog says. “Cave of Forgotten Dreams, you can show it 150 years from now and it still will be completely fresh.” He’s thankful that he doesn’t have to be embarrassed when his grandchildren will eventually see his movies.

Herzog, now 83 years old, appears to live out of step with time. Last fall, on Conan O’Brien’s podcast, Herzog recounted the story of not being able to get his car out of a parking garage in Dublin because he was unable to download the parking garage’s app. But there is a misconception that Herzog is some kind of Luddite. Sure, he still carries a non-smartphone, which he shows me over Zoom, but Herzog emails and video chats with his family around the world. He’s online enough to know the White House has turned a scene from Encounters at the End of the World into a horrible meme.

He is not entirely anti-AI either. He believes we have to be “vigilant,” but sees the way it could create “phenomenal and glorious possibilities in pharmaceuticals, medicine, and mathematics.” As for AI-generated film, Herzog is unimpressed: “All what I’ve seen so far is dead on arrival. Slick and well made, but completely dead. It does not acquire the soul of poetry.”

So many filmmakers — and really anyone in a creative field right now — are reckoning with the encroachment of AI, and there is little consensus of how it will disturb the relationship between artmaking, labor, and commerce. But these conversations are troubled about the immediate future; Herzog, perhaps ever confident from having confronted millennia-old human art in a limestone cave, remains unafraid.

“When you look at Cave of Forgotten Dreams, there’s a deep sense of awe in it, of wonder and mystery. And it has a soul that is not only the filmmaker’s soul. It is a strange soul of human beings who 32,000 years ago created these paintings,” he says, “and AI cannot create this.”

Correction, April 14th: A previous version misstated the name of the IMAX theater in New York. It is AMC Lincoln Square, not Lincoln Center.

The long shot

2025-12-13T11:59:56-05:00

The long take, the unbroken tracking shot, “the oner” — whatever you want to call it, filmmakers agree that it’s one of the most difficult technical achievements in cinema. It’s a feat of creativity, but also great coordination and choreography when a single, tiny mistake can ruin a shot.

Some famous examples: the night club scene of Martin Scorsese’s GoodFellas; more recently, the action sequences in Alfonso Cuarón’s Children of Men and the entirety of Alejandro González Iñárritu’s Birdman. Even a recent episode of The Studio titled “The Oner” — which captures the difficulty of filming a long, unbroken shot — was itself presented as a oner. Seth Rogen’s character calls it “the ultimate cinematic achievement; it’s the perfect marriage of artistry and technicality.” (He then name-checks the three movies I just did, maybe a sign that the continuous single take is something only a certain kind of film bro cares about.)

Yet none of these examples compare to the work of Chinese filmmaker Bi Gan, who is responsible for the most ambitious and impressive long take of the year. His new sci-fi film Resurrection — a labyrinthine expedition through 20th-century China — is capped by an extraordinary 30-minute tracking shot, one that was actually accomplished as a single take. (For comparison, the longest cut in Sam Mendes’ war film 1917 is nine minutes long, each segment stitched together to make the film appear contiguous.) Taking the viewer from nighttime to daybreak, Bi’s oner trails its characters from a violent gang fight on the docks through rainy alleyways to a raucous karaoke bar before returning to the port, where the romance between two leads takes an unexpected, monstrous turn at sea. The effect is dazzling, destabilizing, and unlike anything else you’ll see on screen this year.

What Bi has achieved with Resurrection is an extraordinary technical accomplishment, but also familiar territory for him. His debut feature Kaili Blues culminated in a 41-minute trek up, down, and across the rural mountain town of Dangmai; his last film, Long Day’s Journey into Night, ended with a dreamlike, unbroken 59-minute long take shot in 3D.

Though he spoke through a Mandarin interpreter, Bi, only 36 years old, was extremely talkative and maybe even a bit sly. The long shot might be a tremendous effort of collaboration and coordination across cinematographers, set designers, and a horde of extras, but for Bi, it’s kind of his thing. He’s gotten pretty good at it. He told The Verge how he pulls it off.

This interview has been edited and condensed.

The Verge: You’re strongly associated with the long take. What effect does the long shot have on the viewer?

Bi Gan: For me, I really think about long takes and its impact on the audience. By using long takes, the audience will understand time much better. Because of the fact that when you are watching this in the long duration and because of the mise en scène, that might actually force you to stop thinking about using the kind of jump cuts or fragmented way of telling a story. Now, you are experiencing time in real time, along with the camera.

And when something is a single take, do you want viewers to notice that it’s one long shot?

It doesn’t really matter whether or not they notice that it is a long take, but I do think that most of the audience will. They know that this point of view somehow is moving along with you. The most important is for them to understand time, somehow being uninterrupted because of the long take. but also that time is being compressed in such a way that you experience it as it evolves.

You even play with time in the long take in this film when you use time lapse. Were you trying to warp the experience of time?

In my first film, Kaili Blues, you have three different tenses within that particular long take: something happening in the past, in the present, in the future, but all included in that one long take. Whereas going into my second film, Long Day’s Journey into Night, I really wanted to somehow use this particular long take in such a compressed way to talk about memories. It’s also done in 3D. That brings out something very unique about memories, how some things are very fragmented. But because of the use of the long take, it becomes more holistic and concrete than our actual fragmented memories.

Moving on to Resurrection, I just think that this is the best way to depict that one particular day, which is the last day of the 20th century, going into the 21st century, where two characters elope and then become vampires. One long take for that is because that is the best vehicle to tell that particular story.

Did you know you wanted one of the film’s vignettes to be a long take when you were writing the script, or did you figure that out later?

This particular chapter came first because it’s actually an adaptation of one of my novels titled UFO, which is about two lovers eloping. I do think that when you translate from a written novel to a film, then you need to find the best film language for it.

At first, I was still trying to figure out what would be the best way to sort of make those words come alive in a visual way. It is not until during the shooting process that I had a discussion with my [director of photography] Dong Jingsong and with [production designer] Liu Qiang that we asked, How are we going to do this?

Dong showed me this particular painting by Mark Rothko, an abstract painting with the red colors, with a little bit of other colors. And somehow, the kind of combination of particular colors inspired me to start thinking about using one long, uninterrupted take as a way to tell a story.

We really take this type of film language in a very careful way. We didn’t want to just do it carelessly.

Do you worry that whenever anyone sees one of your films in the future, they’ll expect a long take?

That’s okay, because, you know, I’m not really bound by any rules. You can tell from the films I make.

This movie is in conversation with so much film history. Staying on the long take, are there other filmmakers who do it that you are influenced by?

In terms of this idea of this is very much about a film about films, it’s just on the surface. That is the way that I enter the subject matter of what happened in the 20th century in China.

In terms of the influence from other filmmakers or other films about long takes, I don’t really think that I was informed by them that much. For me, starting with Kaili Blues, the reason why I choose to use a long take was very much on a philosophical level. That was the best way to somehow present, philosophically, what I want to express in that particular film. As a rule, I tend to try to subvert a lot of preconceived notions or concepts about films. So I don’t really see myself as either paying homage or being informed or influenced by other filmmakers’ long takes.

When you start planning out a long take and you’re blocking it, like, what does that even look like? How do you do it?

In terms of the script itself, it’s very much action-based because that’s the only way that you can actually pull this off. One of the biggest challenges for us is to find the right location and space for this story to happen. That’s the start. I was discussing with the art directors how to scout a space that would be the best setting to tell this type of story. We were lucky enough with the team of directors and assistant directors that they looked around and finally found this particular ideal location with railroads, docks, and ports, and then the karaoke bar and the hospital.

Then it’s a constant interaction and collaboration between the teams focusing on the scripts, and then the teams focusing on the sets, and then the teams focusing on technical rehearsals. They have to constantly be working together and evolve in such a way to actually finally make this happen and make it a reality.

The DP and the cinematographer involve the art directors and how we somehow have to work closely and make a lot of adjustments. If during the technical rehearsals, I notice a certain type of technical difficulty that needs to be overcome, then we discuss the possibility of changing or adjusting the script. And then we do another technical rehearsal, and then, finally, finding out that that is the best way to bring everything together.

And then we’ll bring the actors down and do actual rehearsals.

How long is the rehearsal period?

From the moment that art directors are finding space to the actual rehearsals, it took a month.

We could only rehearse at night, and then during the day, the actors and actresses would do some other practices, such as learning to operate a boat and how to sing karaoke.

Are they actually driving the boat?

Oh yeah.

We also have a lot of extras involved for this particular long take. So as a director, I will give them some context — that this is the last night of 1999, and then they will be operating in certain spaces, such as barber shops or karaoke bars. And then based on that, I will ask them to really do character development themselves to think about, “Who you are at this moment of time, in the space, and how would you serve the role as an extra for this particular take?”

In Resurrection, is it all one take or is it stitched together?

In terms of stitching it together, as long as you feel that it is one take, it’s a good thing. And there’s nothing wrong with it. But for my case, for this particular take, it is actually one uninterrupted take.

I was looking for the seams but I couldn’t find anything.

[laughs]

How many times did you have to film it?

Three.

Just three? That’s impressive.

[nods]

Resurrection is in select theaters now.

The ping pong movie is very stressful

2025-11-25T11:31:21-05:00

The origins of table tennis have often obscured its influence — it was literally designed to be the diminutive form of tennis, after all. But the idea of “spin” first originated with ping pong; politically, it would become the thing that opened up negotiations between the US and China under Nixon. There are gestures to this in Marty Supreme, but the new film from Josh Safdie is more interested in the sport as a fixture for outcasts in the ’50s Lower East Side. Loosely based on the true story of Jewish table tennis underdog Marty Reisman, Marty Supreme arrives eager and ready to unnerve you.

“Loosely” is the key word here, unless it turns out the real-life Reisman was a real piece of shit. This version, Marty Mauser (Timothée Chalamet), is as maddening as they come. Anyone who has seen the Safdie brothers’ previous movies Good Time or Uncut Gems will recognize the template: a man who constantly makes selfish choices, every bad decision an excuse to meet a strange character and sprint through the gorgeously shot streets of New York City.

Like Robert Pattinson and Adam Sandler before him, Chalamet is charming but not too charming, driven by his belief that he can become a table tennis champion if he could only scrape together the dough. The film’s second act is possessed by Mauser’s need to acquire the cash for a plane ticket to get him to Tokyo. All his motives are informed by his fixation on beating his Japanese rival Koto Endo (Koto Kawaguchi, who in real life is a Japanese national Deaf table tennis champion). Every transgression that follows is an escalation. Each character — a friend, family member, or lover — is quickly betrayed in pursuit of Marty’s ping pong ambitions.

The Safdies know how to portray this kind of character, and they also know how to draw that performance out of their leading men. They turned Edward Cullen into a dirtbag; they turned Billy Madison into a dirtbag. Paul Atreides? Willy Wonka? The twink from Call Me by Your Name? All pretty good dirtbags too. As Mauser, Chalamet projects a deluded fool, a reluctant father-to-be, and an arrogant hustler. But even through the pencil mustache and thick eyebrows, Chalamet’s charisma grounds the character — or at least prevents him from being too hateable.

A number of other colorful figures populate the film. Gwyneth Paltrow plays a movie star past her days of fame; her asshole tycoon husband is played somewhat unconvincingly by Kevin O’Leary, who Shark Tank fans will know as “Mr. Wonderful.” More impressive surprise turns come from Tyler Okonma (aka Tyler, The Creator) as a fellow ping pong hustler and Abel Ferrara as a funny stereotype of a gangster you might find in an Abel Ferrara movie. These are nice novelties, but the focus is singularly on Chalamet.

As a Safdie movie, Marty Supreme made me crave something new. There’s nothing wrong with a director chasing the same themes or ideas, but this latest outing felt like more stepping backward than forward, or even sideways. (Even this fall’s other “unraveling” character, Josh O’Connor’s much gentler thief in The Mastermind, is a fresher take on an idea the Safdies love.) Chalamet might make a convincing character study of self-rationalization and delusion, but how different is that from Pattinson’s and Sandler’s versions of this? And it doesn’t help that, thanks to the baggy two-and-a-half-hour runtime, this film isn’t as focused or strong as Good Time or Uncut Gems.

In fact, without giving much away, the only thing that really delineates Marty Supreme from this Safdie trilogy of stress tests is its ending — easily the least satisfying act of the film. There are copious scenes of ping pong throughout, but by the time Mauser gets to his final match, the stakes have been deflated by its main character already having suffered a pride-shattering humiliation. The movie maybe should have ended there but instead proceeds on to fulfill the big showdown you might expect from a more traditional sports film. It’s a kinder finale than we’re used to seeing from the Safdies; it’s also a lot cornier.

The Safdie brothers split for their most recent films. Josh directed Marty; Benny directed The Smashing Machine, a portrait of early UFC fighter Mark Kerr, a kind of transformative vehicle for Dwayne “The Rock” Johnson. It’s not hard to see them as similar projects: A24-backed period-specific sports movies starring household names, with expensive needle drops and an odd preoccupation with Japan. They even have similarly unsatisfying endings.

Marty Supreme at least has the effective thriller momentum over the inert Smashing Machine. But they both suffer from the same misguided assumptions. Perhaps it is more compelling to reckon with a sports gambler than an athlete. Maybe it’s easier to judge a character by their vices than by their strengths. But probably of all our contemporary filmmakers, it’s just boring to see the Safdie brothers go soft on their main characters.

Marty Supreme is in theaters December 25, 2025.

The endearing movie that affirms creativity as a human act

2025-11-07T12:20:36-05:00

Peter Hujar’s Day began, as many great works of art do, with a DM. Director Ira Sachs (Passages, The Delta) had just finished reading a recently unearthed interview between the late portrait photographer Peter Hujar and writer Linda Rosenkrantz that took place in 1974. That dialogue — a conversation about creative anxieties, complete with the mundanities of daily life — had been published as a book in 2022.

So Sachs decided to message Rosenkrantz on Instagram about what would eventually become a film adaptation starring Ben Whishaw and Rebecca Hall as Peter and Linda. Deceptively simple and surprisingly moving, Peter Hujar’s Day recreates the interview over the course of a day, set in a single beautiful West Village apartment.

The Verge spoke with Sachs about the challenges of making such a small idea feel expansive and cinematic.

The Verge: You’ve mentioned that the genesis of the film began with you DMing Linda Rosenkrantz on Instagram. What did that look like?

Ira Sachs: I didn’t do any research, so the thing that was really surprising was maybe a month later when I realized she was 89 years old and then I was DMing her. But it was a casual back and forth. She’s super with it, and we’ve become very close in a very touching way — in a way that I think reflects something about her relationship with Peter, actually. Not just that we share this work, but also somehow — I mean, I don’t think I remind her of Peter, but I feel like she reminds me, as the movie does, of the particular nature of heterosexual women and gay men, their friendships. Like, it’s a particular type of friendship that I know really well. And I cherish her.

Is this how you usually start projects? You just, like, DM someone cold?

I start projects with an idea that I feel confident to follow. So, in a way, yes.

At what point did you know this interview would make a good movie?

On the last page. Because I was very moved by the imagery and the feeling that Peter transmitted through his description of 3 in the morning, on the corner of Second Avenue and 12th Street, looking out at the city and listening to the prostitutes on the street below. I felt like that was a cinematic image and a cinematic moment.

And so, the challenge throughout was like, ‘Oh, I need to make that last moment really count.’ All movies, I think, are made in the last moment. And to recognize that, for me, that last moment of the film was both in the moment of 1974 but also filled with loss and melancholy and beauty.

When you say loss, a loss of what?

I could say most simply the loss of that time. But I think, more specifically, I both thought and tried not to think too much about Peter’s death 17 years later from AIDS, that the candle was blown out.

Maybe this is just top of mind, because we’re in biopic season, but what compels you to take such a contained and compact approach to Peter’s life?

Well, I never thought of doing anything else. I wasn’t interested in making a biographical film of Peter Hujar. I was interested in making a film inspired by this particular conversation between Peter and Linda. And what the text had for me was all the intimacy and authenticity that I’m always searching for. Like, in all my work, I just hope to achieve one moment as intimate as Linda and Peter’s conversation.

And because the text is verbatim, it really has the feeling of what it is to spend a long afternoon with a close friend. It also conveys the detail of that time and his life so viscerally — you know, it’s like Proust, really. It really is so densely authentic.

The thing that kind of goes unnoticed about Hujar is he’s an exceptional storyteller. There’s something quite exceptional about his use of language and imagery that I think is quite unique.

The film takes place in a single apartment in the course of one day. But I was really impressed it never really feels claustrophobic. And it also kind of never feels like a stage play either. Peter Hujar’s Day feels like a film. But were you worried about it feeling too small?

I was. The boundaries, the concept at some point — about a month before we started shooting — seemed insurmountable, to be honest. I thought, ‘Uh oh, this was a mistake.’

But liberating myself from the real was really helpful, and also from the physical actuality of the conversation itself, meaning two people across the table talking for an hour and a half. I just decided my version was going to be very different and would instead be 23 scenes over the course of 12 hours.

Constructing this script then, you have all the dialogue already. What was piecing the rest together like?

I spent a couple of weeks with two stand-in actors and my cinematographer, Alex Ashe, on location in an apartment at Westbeth in the West Village, which had been donated. So we had access to this space, and I really spent time photographing these models at different times of day, in different locations. And ultimately, a sequence of those photographs became the kind of guide to how to shoot the film. Really, there was something quite random about what people talked about at certain moments in the film. It wasn’t like I think, ‘Oh, they’re talking about Bette Davis and Joan Crawford. They should be doing that on the bed.’ It was literally like, now there needs to be a cut, because the film needs to maintain an energy.

I took a lot of things away from the film about creativity. But earlier, you said it was a little bit about the things we lost, like this era we lost. How much are you thinking about where this film sits in the modern age or what it’s like watching it today?

I find as an audience member that there is this incredible, unexpected content in the film, which is the window it provides to how hard it is to make art. And that, for me, is something I’m happy to hear any day of the week. I feel it’s an affirming kind of circular conversation that I have as an artist on a regular basis, which is between confidence and doubt. I vacillate very quickly between the two in the same way that Peter questions did he make a good photograph of Allen Ginsberg, or did he make a bad photograph of Allen Ginsberg? And I love that even Peter Hujar — who we now monumentalize and canonize as this great photographer — even Peter Hujar lived with steady doubt at the time.

And for me, that’s very… comforting. It’s really what the impact of the film is in the moment. It’s how it’s received now. This is not a film that nostalgically looks back.

Steady doubt and also worrying about how to make ends meet.

Yes, yes. I think the question of sustainability is one that each of us faces with terror and occasionally hope.

Peter Hujar’s Day is in theaters beginning Friday, November 7th.

The new War of the Worlds movie is even more terrible than you’ve heard but also hilarious

2025-08-13T12:29:53-04:00

There’s a micro-genre of film called “screenlife movies,” where all of the action takes place on a desktop computer. In 2014, Unfriended made a horror movie out of video chat and text messages, and it was successful enough to inspire a sequel. The more thoughtful Searching from 2018, in which a father attempts to find his missing daughter, felt novel and a bit more promising for the genre, even if an hour and a half really tested the amount of time you could look at a scrunched John Cho and Mac OS X.

Amazon Prime Video’s new War of the Worlds reboot is different from those screenlife movies, in that it mostly takes place on Microsoft Windows and is basically unwatchable. There are rumors that it was filmed during the COVID-19 pandemic and shelved because it was so bad. If Steven Spielberg’s 2005 War of the Worlds was a high-budget spectacle, this reboot of the same IP two decades later couldn’t be more opposite: cheap, tacky, and lazy. That said, if you’re a certain kind of viewer — the sort that delights in something that is consistently and unintentionally very funny — it might be worth streaming.

(Some plot spoilers to follow, though “plot” might be generous.)

Ice Cube plays William Radford, a security expert for the US Department of Homeland Security (DHS) with access to the country’s entire surveillance dragnet. (Usually, we refer to characters by their name, but for this article, it feels more appropriate to keep calling him Ice Cube.) The film opens with him logging on to his work computer. Though the DHS is one of the country’s largest employers, with nearly a quarter million people on its payroll, Ice Cube appears to be the only person in the building. In the establishing minutes of the film, he shows off his powers, able to hack into any camera, phone, or drone around the US to spy on people. Menus get right-clicked a lot, as Ice Cube alternates between mundane options and things like commandeer drone.

He uses these huge violations in privacy mostly to violate the privacy of his two kids. His daughter Faith (Iman Benson) is a celebrated biologist, who has engineered something called a “cannibal virus.” Somehow, despite having incredibly deep personal data on every US citizen, Ice Cube actually has no idea what she does for work, and instead, uses his DHS powers to scan what’s in her refrigerator and then calls her on WhatsApp to yell at her for not eating enough protein. Thankfully, in the world of this movie, women love being told what they can and cannot eat. After doling out nutrition advice, Ice Cube checks on his college-age son’s computer hard drive and sees that his other kid has been playing video games. He remotely uninstalls the game, explaining that this will help him focus on school.

A lot of the movie is like this, even once the inevitable alien invasion begins — Ice Cube calling his kids, shouting at them, and then hanging up. What he is supposed to be doing is tracking down a notorious hacker called the Disruptor, a shadow-y anti-government figure that is spouting conspiracy theories about a giant data set of personal information called Goliath. (It is unclear how Goliath is supposed to be worse than the constitutional infringements that Ice Cube is already doing on a daily basis.) But his pursuit of the Disruptor is interrupted by meteors that crash land all across the world. (It’s also unclear why no one sees the meteors coming.)

Aliens begin emerging, and Ice Cube follows along by watching clips on CNN and Fox News in his browser, most of which resembles generic stock footage. When he first sees an alien, he says “goddamn,” followed by an even less convincing “oh, my god.”

It was around this point that I started jotting down all of the apps that Ice Cube has on his computer: Spotify, GitHub, YouTube, WhatsApp, Teams, TeamViewer, Steam, Facebook, Acrobat, Chrome, Zoom, and Ring (though there are probably more that I missed). Along with authentically logo’ed screen elements, using brand-name apps and services lends a sense of realism to the screenlife genre; but in War of the Worlds, it also feels like a lot of product placement, especially for Amazon.

At one point, a character purchases a $1,000 Amazon gift card for a man who is unhoused. Faith’s fiancé Mark (Devon Bostick) is even an Amazon delivery worker. The film’s climax involves Ice Cube ordering a USB drive from Amazon, then having it sent to him by drone, in a dizzying first-person sequence that involves the USB-carrying drone navigating the DHS building. The logo for Amazon’s “Prime Air” is visible during this whole scene. (Among all the flagrant product placement, there was a notable absence of Alexa devices — likely because it would suggest to the viewer that Homeland Security worker Ice Cube could spy on them through it.)

It turns out the aliens are hungry for “data.” Read that previous sentence as literally as possible. The aliens actually eat data, which the film repeatedly calls “our most precious resource.” In case it is not clear enough what is happening, the movie reemphasizes this by declaring that “all systems have been data drained” and that “every system has been penetrated and drained of data.” Watching the aliens devour a government database, Ice Cube says, “They’re in hyper download. They’re being emptied.” Things get personal when he opens the Facebook page of his dead wife and watches as all the images on her profile begin disappearing, with him holding back choked tears while I was trying not to choke on my seltzer from laughter.

I don’t often indulge in a “so bad it’s good” recommendation, but if the details I’ve laid out so far are funny to you, then there are plenty more of them crammed into this 90-minute movie. The density of bad dialogue, plot holes, and shameless plugs in this dreck could rival a neutron star.

I’ve seen reviewers call this the future of movies, one where tech companies generate huge ads for themselves. Though I imagine Amazon will take more care with its Denis Villeneuve-helmed James Bond film, there are already rumors circulating that Jeff Bezos is pushing for his wife, Lauren Sanchez, to be Bond girl. But if it’s just as easy to watch War of the Worlds and see it as an oblivious, self-condemnation of Amazon itself. And it’s not just critics that feel this way. The movie has a dismal 2.7 user rating on IMDb, a site also owned by Amazon. If War of the Worlds represents anything, it’s a strong rebuke — a signal that viewers know garbage when they see it.

But if you wanted to watch a good movie, the 2005 War of the Worlds holds up pretty well, too.

40 years later, Brazil is as prescient as ever

2025-08-05T17:35:44-04:00

Brazil opens with a bureaucratic error. A fly gets stuck in a typewriter, changing the surname of Archibald Tuttle to Archibald Buttle, a misprint on a form that dictates the government forcibly detain a suspected terrorist (Tuttle) but instead leads to the arrest of an entirely innocent man (Buttle). If the inciting events of our great science fiction films have been hostile aliens, seductive robots, and reckless technologies, Terry Gilliam begins his with a humble typo.

Rewatching Brazil in 2025 — nearly four decades after its release — it’s hard to understate how well this movie holds up. Wildly inventive at every turn, Gilliam’s satirical vision of a cruel and violent bureaucracy rings eerily true of this political moment. The film finishes a weeklong run at New York’s Film Forum with a new 4K restoration, which you can also get on Blu-ray. (And honestly, the non-4K version of Brazil that you can perennially stream on The Criterion Channel still looks great too.)

A lot of that has to do with Gilliam’s hysterical dystopia — Mad Men by way of Wolfenstein. Brazil also imagines a hyper-efficient future that never made the leap to digital. Pneumatic tubes shoot paperwork between offices; seas of typists clack forward the cogs of an industrial machine. Everything in this world is an Orwellian/Kafkaesque melange of forms and stamps and obtuse processes.
The experience of watching Brazil is at once being impressed by how it looks while also being horrified by what’s depicted. The ominous cityscapes have wonderfully art deco touches, yet the gargantuan buildings cast long, haunting shadows; many of the sets take inspiration from Nazi iconography, complete with gigantic eagles and massive lobbies guarded by stormtroopers. Also, look at this logo:

Isn’t that the best movie logo you’ve ever seen?

A perfectly cast Jonathan Pryce inhabits Sam Lowry, a mid-level bureaucrat. He lives in a small apartment, complete with dysfunctional Rube Goldberg gadgetry that ends up pouring coffee on his toast. (The film has no shortage of Gilliam’s adoration of slapstick, a carryover of his Monty Python days.) Lowry’s mother and friends push him to be more ambitious. Yet he resists the rat race, turning down a promotion to a much more prestigious branch of the government simply because he isn’t interested. In this dystopian world, oppressed by the hierarchical structures of capitalism, the only hero is a slouch.

A fantastical/horny dream plotline is the most Lowry gets activated, and as he chases down the culprit for the Tuttle and Buttle mix-up, he encounters several different departments foisting the blame off on other offices. “Information Transit got the wrong man. I got the right man,” says one bureaucrat. “The wrong one was delivered to me as the right man; I accepted him on good faith as the right man. Was I wrong?” There is no accountability in this government, and characters act with self-interested careerism in mind over any semblance of morality. After Buttle is killed, Lowry has to deliver a receipt to his widow.

Earlier this year, as part of the Trump administration’s attempt to deport undocumented immigrants, ICE illegally deported Kilmar Abrego Garcia. He was sent to El Salvador before the government admitted it had made an error. Then the agency backtracked, claiming it had never made a mistake. In response to calls to return Garcia to the US, the Department of Homeland Security claimed it had no authority to do so.

The deflection of responsibility, the ludicrous reasoning, the deferential loyalty to the state — these are the things Terry Gilliam satirized in Brazil. Most science fiction films emphasize the dangers of technology; Gilliam saw the sinister machinations of bureaucracy. Watching Brazil 40 years later, it’s even clearer what we were being warned about. Some of that clarity is literally the 4K restoration. But even through all of Gilliam’s gags and elaborate sets, we see all the twisted incentives that eventually normalize fascism.

Eddington gets the pandemic right but still isn’t a great movie

2025-07-16T16:28:32-04:00

A24 is known for its prestige arthouse films, but in its early days as a distributor, it made most of its money from elevated horror films like Ari Aster’s Hereditary and Midsommar. Over a decade in, the ambitions of A24 and Aster have expanded beyond genre film. But for both, the more recent results have been mixed.

Eddington, Aster’s latest, feels like a continuation of the maximalist guilt-trip Beau Is Afraid. Joaquin Phoenix stars once again, though the concerns here are less Jewish and Oedipal and more wokeness and conspiracy theories. It’s grounded in the contemporary: the early days of the COVID-19 pandemic, specifically. The movie’s tagline is “hindsight is 2020,” which is fitting for a movie that is clever and empty.

In a fictional New Mexican town, Sheriff Joe Cross (Phoenix) is unhappy about statewide COVID-19 precautions. He’s not a clean stand-in for any particular kind of conservative; he simply hates wearing a mask, claiming that a KN95 is oppressive as someone who suffers from asthma. Though coronavirus has not arrived in Eddington yet, the anxiety created by mask mandates, grocery lines, and online misinformation loom just as large over the town as the virus itself. Cross doesn’t like what the world has become so he announces on a livestream that he will run for mayor.

Eddington is supposed to be a microcosm of what’s wrong with the US, though Aster’s diagnosis feels overly broad. The film works better when it comes to the finer details. The town is full of weirdos, embodied by a strong and familiar cast: Pedro Pascal plays incumbent mayor Ted Garcia, who’s in the middle of selling out Eddington’s future with the construction of a massive AI data center; Emma Stone plays Cross’ wife Louise, who makes scary dolls as a hobby; Deirdre O’Connell is her mother Dawn, a YouTube-susceptible conspiracy theorist; Austin Butler makes an appearance as Vernon, a cult leader with tattoos that evoke a Hillsong pastor. In some ways, watching the first hour of Eddington feels a bit like watching a Wes Anderson movie or maybe Yorgos Lanthimos’ cruel triptych Kinds of Kindness. The delights come from watching a bunch of recognizable actors inhabit and bounce off odd characters. (You get the sense that they had a good time on set, too.)

As a time capsule of 2020, the movie also confronts Black Lives Matter protests, though the leader of Eddington’s anti-racist movement is a white high schooler (Amélie Hoeferle) who repeatedly gives her ex, a Black police officer named Michael (Micheal Ward), tone-deaf lectures about joining the marches. Two other teenage characters begin protesting mostly because they think she’s hot. Aster finds a lot of his jokes in the grating nature of social justice language.

For all its themes, the early parts of Eddington are light on moralizing or righteousness. Even if the setups are somewhat obvious — the annoying performativeness of the left, the boneheaded ignorance of the right — the punchlines mostly land. Eddington posits that the thing that both sides can agree on is that, progressive or conservative, we are all manipulated by our phones and the incentives of social media. If the idea is obvious, at least Aster pulls it off convincingly.

That is, until the last hour or so. Maybe Aster just struggles with endings? Tonally, Hereditary’s ending works as a de-escalation, though the final twist is unsatisfying; the atrocious last act of Beau Is Afraid is as surreal as it is irritating, with Richard Kind stepping in as a kind of inverse Bing Bong. In Eddington, though, all of the threads that Aster puts into place seem to unravel as Joe Cross’ motivations take an unconvincing, violent turn. Without giving it away, the plot moves in an absurdist direction — a fine choice for a black comedy, but a disappointing one in a film that begins with a more compelling, grounded worldview.

Maybe five years isn’t enough time to understand what exactly the pandemic did to us as individuals or as a society, but I think anyone would suspect that it’s more complicated than “it broke our brains.” If nothing else, Eddington proves that that reasoning is deeply boring on a narrative level. Joe Cross’ unexpected arc, despite Phoenix’s beguiling performance, comes across as confused and unearned.

Eddington is memorable, though. Again, it’s the details. At the police station, an officer’s desktop computer is covered in anime stickers; you catch a brief glimpse of a TikTok video of a white woman doing a celebratory dance after reading Giovanni’s Room. I could be convinced to watch Eddington again just to see all the hilarious, meticulous touches that Aster has embedded in the scenery. As a filmmaker, he demonstrates a strong care for the craft — so much thought has been put into the cinematography, the sets, the evocation of the pandemic. You just wish that same effort had been put into any of the film’s ideas.

Eddington is in theaters nationwide July 18.

Tall Tales is a critique of AI — so why do people think it was made with AI?

2025-05-09T10:19:49-04:00

For the past three decades, multidisciplinary artist Jonathan Zawada has produced art across various mediums: sculptures, paintings, videos, installations, and more. He’s also been making visualizations with his friend Mark Pritchard, a record producer and experimental musician. They became close friends, talking nearly every day. But the relationship changed when Pritchard brought him a new, daunting project: a collaborative album with Thom Yorke, the frontman of Radiohead.

“I’m a massive fan, and I’ve found he’s kind of a terrifying person,” Zawada says. But despite being intimidated, he found Yorke to be a great collaborator. “He’s so switched on, and he’s so clear and concise… Yeah, I was very nervous a lot of the time.”

Before the record, Tall Tales, was even a complete album, Zawada was working on a visual accompaniment to it. Songs would arrive in his inbox — oftentimes just sketches or demos — and Zawada would send back pictures or start a collaborative whiteboard. It’s just whatever popped into his mind after listening to the music: Dutch painters Pieter Bruegel and Hieronymus Bosch came up a lot. Five years later, the album is now available and Zawada’s accompanying “visual experience” is in select theaters today.

You can see the influence of Bruegel and Bosch in Tall Tales, though twisted into something modern and digital and absurd. The movie features a carnival of eerie, unnerving monsters set against mesmerizing, technicolor landscapes. There are interludes of real-life footage, but even that is used to destabilize the viewer with the vantage point of drones and surveillance cameras observing the massive scale of construction sites and global shipping apparatuses. Tall Tales gestures to the nightmare of contemporary life, but it is also tremendously funny: contorted CGI townspeople dancing to prickly electronica, grotesque mutated heads marching along to Thom Yorke’s anxious falsetto. Between songs, the movie cuts to a video-game-inspired world map, where a little bird walks the viewer to the next stage/song. (At the screening where I saw Tall Tales, the audience laughed at this, without fail, each time.)

Zawada’s work very much explores the line between artificiality and humanity, and he’s thinking often about the ethical and aesthetic ramifications of technology. Which is why it was especially painful when one of the early videos from Tall Tales dropped and people started accusing the movie of being made with AI.

Zawada doesn’t actually read the comments, but Pritchard does. He was wounded to find out that people believed AI had been used on the video for “Gangsters” when it had not. “[Pritchard] tells me about all this grief that it gets from people assuming it’s made with AI, and then bitching about it and complaining that it’s made with AI, even though none of these videos have been made with AI,” Zawada says, of the videos that had been released.

Perhaps it was Zawada’s aesthetic — one that expresses a malformed, distorted version of its CGI influences — that aroused viewers’ suspicions. The easiest red flag that an image was generated with AI is to count the number of fingers, yet a creature with too many fingers would feel right at home in the universe of Tall Tales.

“If it looks like something that’s not real now, people think it’s AI,” Zawada says.

Still, he understands why fans are so allergic to anything that even has a whiff of AI. Particularly in music, an industry that has contracted and made livable wages for artists extraordinarily rare, the conversations around AI, creativity, and labor become heated quickly. (Recently, Stereolab released a music video made entirely with AI, much to the chagrin of fans. The most upvoted comment on the YouTube video: “we love stereolab, we hate AI.”)

Like a lot of artists, Zawada was already worried about AI. He’s not against using it in some cases — he admits that he deployed it in some places in Tall Tales, like a few environmental backgrounds and the texture for a fish body — but all his usage comes from local installs of publicly available models rather than Midjourney or Dall-E. (Some would argue that what Zawada is doing doesn’t even constitute generative AI these days.)

But his concerns about AI are more existential, and predate this moment. Will anyone want to make art in a world where the attention economy is so consumed by a flood of content? He recalls a time before the internet and social media. “If you wanted to get people to pay attention to you, you’d write a song,” he says. “And people listen to you and that feels good. Then there’s sort of like a flywheel feedback loop of how culture gets made and how art gets produced.” That’s all changed now. “You don’t need to make a piece of art. You just need to make a social media post.” The merits and value of what AI produces continues to be debatable, but no one will deny that it generates things at a scale that historically seemed unfathomable.

One of the most memorable visuals accompanies the song “A Fake in a Faker’s World,” which features a seemingly endless row of mechanical arms painting an ever-changing landscape — a not-so-subtle depiction of robots making art. Eventually, it cuts to a human’s face, an expression melting into a rainbow, before returning to the robotic painters, now removed from their environment and floating as a careening kaleidoscope of machines reproducing the same hideous image to infinity. The drums kick in, Yorke begins to croon, and finally the arms are creating unique images.

Thematically, Tall Tales is a bit of a cipher, but as a project, it began with Zawada thinking about forgeries. He found AI art more compelling in its early days, when it generated things that looked insane. If the internet has a primary characteristic, it’s that things can be copied effortlessly and endlessly. As AI gets harder to detect, maybe we’ll be drawn to the things that are obviously fake.

The Tall Tales album is available now, and the film is screening in select theaters around the world today.

Box art

2025-05-05T11:01:31-04:00

Literary journals might have a stuffy reputation. But since its conception in 1998 by author Dave Eggers, McSweeney’s Quarterly has been anything but, opting instead to be an endlessly mutating delivery system for writing and art. It has been a hardcover book, a paperback, a newspaper. Once it was a bundle of mail; another time, a deck of playing cards. Imagination and capriciousness have defined McSweeney’s for nearly three decades.

The latest issue, edited by Rita Bullwinkel and guest-edited by two celebrated writers — cartoonist Thi Bui and novelist Vu Tran — attempts to capture the messy and disparate nature of the Vietnamese diaspora with a package that is, by design, messy and disparate. The 78th issue of McSweeney’s, “The Make-Believers,” arrives in a cigar box with painted illustrations by Bui containing several unique booklets of stories, essays, and illustrations that try to pin down the elusive trappings of Vietnamese identity.

Curious about the tremendous effort of putting together such a unique package, The Verge spoke with Bui, Tran, and art director Sunra Thompson about how “The Make-Believers” came together. It turned out that although it might sound like this issue was planned to coincide with the 50th anniversary of the end of the Vietnam War, that was actually a coincidence. In the spirit of McSweeney’s — and perhaps any ambitious creative project — it was equal parts hard work, serendipity, and chaos.

How did the project come to be?

Thi Bui: It was conceived on a hilltop in Marin County. I was taking a hike with Dave Eggers, who became a friend after we worked on a screenplay together. That movie is never going to get made but we have a friendship out of it. Occasionally we’ll take hikes and talk about art and life.

I had just come back from DVAN, this incredible residency in the south of France with these other Vietnamese writers. This was one of those really special experiences where everyone fell in love with each other, and it was highly productive and magical. So I was just trying to describe that to Dave.

I think he’s always like canvassing his brain for how to uplift people. Out of the blue, he was like, ‘Do you guys want to take over an issue with McSweeney’s like one of you could be the guest editor? You know, it wouldn’t be that much work.’ I took the idea back to the group and only Vu really knew that much about McSweeney’s.

Vu Tran: I remember when McSweeney’s first came out. Back then, if you pitched them a story, they sent you rejections — little slips of paper as rejections. I still have my six or seven rejections. That was like 20 years ago or longer. And it’s so funny. I would have never thought that the way I would actually get into the magazine is to guest-edit it.

From the jump. Did you guys know you wanted to do an issue timed to the 50th anniversary of the fall of Saigon?

TB: No, we were kind of flying by the seat of our pants with that one.

“I had all these amazing enablers.”

VT: The issue just happened to coincide with a spring 2025 publication date. It was completely unplanned.

Sunra Thompson: The way they aligned was kind of by accident.

VT: It was a great coincidence.

What were the early ideas for the issue like? Was it always a cigar box?

TB: It was always a box. I wanted something really nice that’s evocative of old Vietnam. It would be some sort of treasure box. Other people were like, ‘But if it’s too expensive, no one will be able to afford it. And then they’ll be afraid to open it, because it’s too fancy. You’re so scarred in your refugee-ness.

VT: Yeah, yeah, it was very Vietnamese. When Thi described it to me, it just implicitly felt right: that kind of nostalgia for a beautiful past, but also an understanding that history is filled with all this other stuff that’s not quite as elegant and and not quite as neat. So it’s elegant and weird.

ST: Yeah, Thi really did have a very clear vision of what this would look like, which was especially helpful for an issue like this that has a lot of different components. We’ve done issues that come in a box or something, but for each different component, you need a different cover design.

But Thi knew what every cover was going to look like right away. She had artists in mind for each of these covers, too. Her vision of this issue made a lot of these decisions pretty quick.

TB: I think this is why it just felt like such a dream project because all I had to do is imagine it. And then I had all these amazing enablers.

As the designer, what was your first impression after being told about this idea with all these different components?

ST: For the Quarterly, the idea is that every issue is packaged in some unique way. So if I go into a stationery store, and I see some weird notebook or something, I will sometimes just take a picture and ask a printer like, ‘Can you make this thing?’ That’s a big part of my job. How can I package a book in a different way? So usually when an editor or an artist has a packaging idea, I just immediately email a printer to see what they can make.

It’s kind of my favorite part about doing projects. At the beginning, when you’re just asking printers to make dummies — it’s just pure potential.

TB: It’s such a different experience to get to work with a publisher who says yes. This was such an incredible opportunity to keep making the project weirder and fancier.

Were there production challenges?

ST: On the printer side, there were some pretty run-of-the-mill issues. For example, there’s a lot of foil. The first sample I got, the foils misregistered with the ink of the lettering. I got spooked, and I just abandoned that idea.

TB: I think we had one idea that we couldn’t execute, which was like having different kinds of paper stock in the same bound book.

“I think deadlines are nice sometimes, because it forces you to kill your own dreams.”

ST: I forgot about that. Like, smash together different aesthetics.

TB: A section of the menu called classifieds, and we were trying to print it on newsprint. But I think we maybe ran out of steam at that point.

ST: That happens with projects that are very complex. You do sort of have to choose the things you want to focus on.

TB: Yeah, at that point I was like, ‘I wanna preserve Sunra’s mental health.’

ST: Yeah, it’s true. I appreciate that. I start a lot of projects thinking as extravagantly as possible. And then you’re kind of in the middle of it, you become a little less precious when you realize, it’s a periodical, too. We have to get four out a year. I think deadlines are nice sometimes, because it forces you to kill your own dreams, which can be like a nice lesson.

When I worked in a print magazine, we would joke that the print process would make everything better and better. And like, as you approach the close, you just made everything 10% worse to just get it done.

ST: It’s so true. But it’s probably good. I probably would never get anything done without a deadline.

What about the challenges on the editing side?

VT: For me, the most educational and fascinating aspect of this particular project was the translation. We had to get someone to proofread the Vietnamese right?

My favorite, but also the most difficult experience editing for me, was Doan Bui, who writes in English. But English is her third language. And she writes in a very vibrant voice, but it’s not grammatical most of the time and it’s repetitive. It became too time-consuming to constantly ask her, ‘How should we change it?’ So we just had an agreement. She said, ‘You make the corrections for me.’

I found myself in some cases rewriting things in a way, like translating her voice from an imperfect English into a clear, more engaging English that also captured my sense of her intended meaning, but also her: the foreignness, and the personality in her voice.

It became this really interesting thing that ended up reinforcing this idea Thi had about our shared imagination of what it means to be Vietnamese. It ended up reinforcing the themes of the issue, which I just found really satisfying, even though it took like a chunk out of my life.

TB: Yeah, I definitely got like a new wrinkle in my brain — and maybe on my forehead, too. I was just thinking about the allegory of the three blind men and the elephant? And I think that’s us with Vietnamese culture and language. We each know so little, and sometimes have a completely different interpretation of a word or an idea.

Sometimes we’d be like, ‘Wait! Doesn’t it mean this? Wait! What is that?’ And we would call our parents to confirm something.

I feel like when I call my parents to explain a thing or a word or phrase, they just end up disagreeing.

TB: Everyone’s an unreliable narrator.

For Scale

2025-04-23T12:37:44-04:00

Operation Rolling Thunder was meant to be an act of persuasion. The US believed that a drawn-out bombardment would pressure the North to cease its aggression on the South — or, at least, encourage it to ease up. “I saw our bombs as my political resources for negotiating a peace,” President Lyndon Johnson claimed. His framing was belied by the words of Air Force Chief of Staff Gen. Curtis E. LeMay, who said, “We’re going to bomb them back into the Stone Age.”

Rolling Thunder was supposed to take eight weeks. Instead, it lasted more than three and a half years, with hundreds of thousands of sorties. It was longer than any bombing campaign during World War II or any other war that came before; it remains the longest bombing campaign in history. It cost the US $900 million, compared to an estimated $300 million in damage to the North Vietnamese. Given that the conflict continued for another seven years, it’s safe to say that Rolling Thunder was not very persuasive.

Still, Secretary of Defense Robert McNamara thought it could work. Before the war, McNamara had made the unusual choice of leaving his position as the president of Ford Motor Company — at the time, one of the highest-paying jobs in the world — to work for the White House. He was a numbers guy, who believed that everything could be solved through efficiency metrics. The method was called “scientific management.” That belief in quantification had boosted American corporations; certainly it could do the same for US foreign policy.

Rolling Thunder reflected McNamara’s ambitions and approach toward Vietnam. According to a biography by Deborah Shapley, he saw the bombing campaign “as a balance sheet, with the number of enemy targets hit in one column and measures of enemy activity in the South on the other.” Not coincidentally, many historians see Rolling Thunder as a microcosm for the conflict itself — the hubris of the United States, its inability to understand what kind of war it was getting into. But even if one looked at the strategy on McNamara’s terms, they would see that the numbers offer a glimpse into the size and scope of what true failure looks like.

During Rolling Thunder, between 1965 to 1968, the US dropped 864,000 tons of explosives over Vietnam. We know this precise measurement because the military keeps active and accurate records, which it did for allocations, accountability, and so McNamara could inform policy decisions. For scale, the RMS Titanic weighed about 58,587 tons. The munitions dropped during Rolling Thunder would be the equivalent of nearly 15 Titanics.

But it’s a bit hard to imagine how big a Titanic is, let alone 15 of them. It’s easier to conjure a modern Ford F-150 pickup truck, the country’s most popular automobile, which weighs around 5,000 pounds on average. So imagine the bombs dropped during Rolling Thunder as nearly 344,000 pickup trucks — the kind you’re most likely to see on the road, but hundreds of thousands of them. For context, your average Ikea parking lot has the space for 1,700 automobiles. So envision about 202 Ikea parking lots, completely filled with pickup trucks.

Though Rolling Thunder was primarily a bombing campaign, it was also an early opportunity for the US to flex its air combat superiority. The US deployed variations of an explosive projectile developed by Raytheon, known as the Sparrow, for plane-to-plane encounters. It is now infamous for being a terrible missile — accuracy is an efficiency metric, and the Sparrows were not accurate. Military studies conducted after the war found that only 9.2 percent of Sparrows fired during the war hit their targets. A whopping 66 percent of them malfunctioned; the remaining failures just missed. The batting average of one of the worst hitters of all time, dating back to the beginning of Major League Baseball, belonged to Bill Bergen. He played for the Brooklyn Dodgers around the turn of the 20th century and batted about 0.170 — so about two times better than a Sparrow missile. There were plenty of excuses for the Sparrow’s performance: poor training, poor production, poor maintenance. It didn’t change the fact that each missile may have cost as much as $225,000, which, after inflation, would be $2.3 million a piece today.

But McNamara’s favorite efficiency metric was “loss exchange ratio.” It is the simple math of determining the quantitative relationship between how many you lost to how many they lost. That figure asks: what is the value of a life? You could determine the average price of saving a life to determine a human being’s worth. Conversely, as the military does, you can calculate how much it costs to kill them. The formula is straightforward: how much you spent divided by the number of deaths. If this sounds blunt, it is exactly the equation the US used. Loss exchange ratio is better known as kill ratio.

Being a man of “cool efficiency,” as he called himself, McNamara had an advantage in continuing to push through these doomed plans. One aide described him as being forceful and convincing. In meetings, McNamara arrived with “briefs, numbers, ratios, estimates, and projections.” (The same aide also described him as “exhausting.”) Gen. Maxwell Taylor, chairman of the joint chiefs of staff, described encounters with the defense secretary as being “bombarded.”

It’s why the explosives kept falling, even as the numbers didn’t look great for Rolling Thunder. The most generous estimate of casualties claims that 21,000 enemy combatants were killed, meaning that, after spending $900 million, each one cost the US around $42,857. (Adjusted for inflation, that’s $9.2 billion — so $438,095 per life. Again, not very efficient.) McNamara considered body count to be the most precise, objective metric for success, but at no point did that factor in the more than 182,000 civilians killed during Rolling Thunder.

That was just during the three-year span of Rolling Thunder. Over the course of the two decades the US military was in Vietnam, the US dropped an estimated 5 million tons of explosives. That’s twice as much as during the entirety of World War II, and it remains, to this day, the largest bombardment of any single country ever.

Five million tons of bombs, or if it’s easier to imagine, 85 Titanics.

left | https://s3.us-east-1.amazonaws.com/assets.sbnation.com/csk/uploads/verge-features/american-war/spot-illos/spot_1.jpeg | The Vietnam Veterans Memorial stretches for 400 feet. One built for non-Americans lost with the same density of names would stretch for nearly five miles

The best way to honor the dead was with a competition. The Vietnam Veterans Memorial Fund asked, who could design the best war memorial? The requirements were sparse: it must use the names of the lost soldiers; it must be “reflective” and “harmonious” while making “no political statement of the war.” The competition opened in the fall of 1980, and submissions were blind judged. Every entry was given a number rather than a name. There were 1,421 proposals, and a jury of eight unanimously chose the winner: entry #1026.

Maya Lin was 21 years old and studying architecture as an undergraduate at Yale University. She’d already been thinking about death. Earlier, for a class, she’d designed a memorial for an imagined World War III; she turned in a drawing of an underground tomb, a concept that deeply upset her professor. For the Vietnam Veterans Memorial competition, she sketched a cut in the earth; a sunken, black stone listing the soldiers’ names and nothing more. “The need for the names to be on the memorial would become the memorial; there was no need to embellish the design further,” she said.

Lin was surprised to win the competition. She’d submitted it in a college class and received a B. (No matter, her professor had entered the competition and lost.) Being entry #1026 had obfuscated the fact that she was Chinese American from the judges, but once her proposal was announced to the public, there were concerns that the memorial should not be designed by someone of Asian descent. Lin spent several tortuous months in Washington, DC, overseeing the project, enduring criticism of her design from all sides. She recalls one Washington Post op-ed dubbing her work “an Asian memorial for an Asian war.” (She was born in Ohio.)

As Lin’s work moved through a bureaucratic approval process, other design choices were called into question. One crucial facet of Lin’s idea was to list the names chronologically; veterans groups resisted the idea, saying it would be difficult for visitors to find the exact location of where a soldier was honored. Wouldn’t it be so much easier to list names alphabetically? But Lin fought hard to preserve the chronology, and she prevailed in the end: an honest accounting of death over a convenient one. The memorial would live firmly in time, rather than outside of it.

There was pushback on the color, as well. “One needs no artistic education to see this design for what it is: a black trench that scars the Mall. Black walls, the universal color of shame and sorrow and degradation,” said veteran Tom Carhart. In an essay for the New York Review of Books many years later, Lin defended her choice. “I do not think I thought of the color black as a color, more as the idea of a dark mirror into a shadowed mirrored image of the space, a space we cannot enter and from which the names separate us, an interface between the world of the living and the world of the dead,” she wrote. The prompt had asked for “reflective” — the black granite was quite literally reflective.

Unconvinced, Carhart and other critics suggested the wall be made white and adorned with a more conventional eight-foot-tall statue of wounded soldiers. Plus, they wanted a flag right in the center. Lin objected, claiming the additions violated the integrity of the work. The US Commission of Fine Arts, which had final say, heard arguments in favor of and against Lin’s design, and eventually settled on a compromise: Lin’s vision would remain intact, but a statue and a flag would be added — not in the center, but off to the side. No one informed Lin of the additions, and only after reading about it in the paper did she learn her vision would be undermined. (“They didn’t have the stomach to tell me,” she said.) The memorial was completed and dedicated in November 1982, but by that point, Lin had already left Washington.

In a city that is full of bright white neoclassical statues and monuments, the Vietnam Veterans Memorial is an unabashed piece of the land art movement, appearing almost like a dark gash carved into the ground. Lin even said she wanted it to look like a “wound that is closed and healing,” but the fact remains that black granite is a static material, a hard rock that is as close to permanence as we have on this planet.

The names set in the stone are cast in the typeface Optima. (Decades later, John McCain would deploy the same type in his presidential campaign logo.) Every name on the memorial is the same size, giving equal weight to each life, regardless of military rank. There are 58,395 names in total, representing the soldiers that were killed or missing in action from 1956 to 1975. For scale, if you met an average of two new people a day, every day — an incredible social clip — you would encounter only 55,518 people, assuming you lived to the American average age of 76. More than a lifetime’s worth lost, now memorialized as a small name chiseled into a slab of granite.

That death toll has become a strange marker to convey magnitudes of loss: for traffic accidents, gun violence, and other wars. During the pandemic, several outlets noted when the number of people killed by COVID-19 surpassed the fatalities of US soldiers during the Vietnam War. This is perhaps the legacy of Robert McNamara: an emphasis on body count, the metrics-driven approach to understanding death.

But even just looking at the numbers, there is the erasure of a greater figure: the 3.8 million Vietnamese soldiers and civilians that were killed during the war. That’s roughly the current population of Berlin or Los Angeles.

McNamara used the kill ratio as the key metric for the war, guiding many of his policy recommendations. The 58,000 Americans killed compared to the 3.8 million Vietnamese killed brings the kill ratio to a staggering 1 to 65.

Maybe it’s easier to imagine that ratio in other terms: a double espresso shot compared to a gallon of milk; the Scoville difference between a common serrano chile and a ghost pepper.

The Vietnam Veterans Memorial stretches for 400 feet. One built for non-Americans lost with the same density of names would stretch for nearly five miles.

right | https://s3.us-east-1.amazonaws.com/assets.sbnation.com/csk/uploads/verge-features/american-war/spot-illos/spot_2.jpeg | Being impact-free is not the same as a total absence of bombs: Vietnam will never be close to completely clear

The remnants of the war are scattered everywhere across Vietnam. This manifests, in the most literal sense, as unexploded ordnance. These leftover explosives are still littered across the entire country. The Boeing B-52 Stratofortress could carry up to 108 bombs, each of which would then disperse as many as 600 tennis ball-sized “bomblets,” destroying everything in an area that was one mile long and half a mile wide. When one of McNamara’s deputies asked why the US deployed B-52s — a plane famous for its devastating power and lack of precision — Gen. William E. DePuy delivered the answer calmly and honestly: “because they’re there.”

This technique is called “carpet bombing” because it affects a large area, the way a carpet might cover a floor. The most famous ones were during World War II: Dresden, Hamburg, Tokyo. As recently as 2023, the US has controversially sold cluster bombs to Ukraine. But their usage was never more intense than they were over the 43 square miles of Quảng Trị, a rural province in Vietnam that was so thoroughly leveled that only 11 of its 3,500 villages were left alone by 1975. Quảng Trị has been dubbed “the most bombed place on Earth.”

While cluster bombs are an efficient way to annihilate large areas of land, the adorably named bomblets have a high failure rate as high as 30 percent. After the war, millions of dud cluster bomblets remain scattered across the country. Since the war ended in 1975, they have killed or injured more than 100,000 people. Estimates indicate that 17 percent of the entire country is still contaminated by leftover explosives — millions and millions of more bombs.

Several nonprofits, like Project Renew and the Mines Advisory Group (MAG), are dedicated to ordnance removal. They employ locals to survey large, often forested areas; teams of deminers locate explosives with off-the-shelf metal detectors. The work pays the equivalent of $500 a month, which is more than double Vietnam’s minimum wage. A team of 14 clears approximately 38,750 square feet a day — about half a professional soccer field. At that rate, it takes almost a year of sustained work to clear the area of a single cluster bomb. One of the hundred dropped from a B-52 would have taken about 30 seconds to reach the ground, and decades later, it would require more than 40,000 hours of human labor to clean up.

While it may sound like a dangerous job, heavy training and stringent safety precautions have resulted in very few accidents or injuries. At the end of each day, the unexploded ordnance are gathered and safely exploded. Project Renew says it has detonated more than 815,000 of them so far, while MAG has detonated another 400,000. The work in Quảng Trị, where the problem is the worst, hopes to be entirely clear by 2035, 60 years after the end of the war.

Still, as of 2024, at least half a million hectares of land in Vietnam have been cleared. The remaining area that needs to be cleared is another 6.6 million hectares. That means after half a century, only 7.6 percent of the contaminated areas have been deemed safe and ordnance-free.

At least one estimate suggests that it will be another hundred years of sustained work before Vietnam, Laos, and Cambodia can be decontaminated of explosives; doing that math is hard, because it depends on so many variables, including the consistency of funding. The US has given $750 million for the cleanup effort, which seems like a large sum until you realize that the country spent $352 billion ($2.2 trillion after inflation) on the war effort. Earlier this year, the Trump administration suspended funding for bomb removal in Vietnam. Given the size of the issue, and how much progress has been made in five decades, it’s difficult to imagine a bomb-free Vietnam in the next 500 years — unless the current pace is significantly accelerated. The goal, according to a MAG representative, is to be “impact free” — that is, land safe enough to be developed, for communities and economies to flourish. Being impact-free is not the same as a total absence of bombs: Vietnam will never be close to completely clear.

When something is so big, it tends to become abstracted, simply so our minds can grasp them. This is normal. We abstract many of the things that are important to us: money, time, life. We only imagine things — value and worth — in relation to other things. What is 7.6 percent? That would be equivalent to running a marathon and quitting after the second mile. Or starting a two-hour-long horror movie and deciding it’s too scary less than 10 minutes in. Or living to kindergarten age in an average human lifespan.

I say all this, of course, just for an idea of proportion, for a sense of scale.