Lost in the Funhouse: A Conversation with Guy Maddin and Evan Johnson on The Forbidden Room and Other Stories

By Mark Peranson

Originally published in Cinema Scope 63 (Summer 2015).

Like being sloppily slapped by a wet salmon to the point of submission, such is the impact of Guy Maddin and Evan Johnson’s inventive, audacious, and outright hilarious tour de force whatzit. Sure to ride roughshod over numerous territories following its premieres in Sundance and the Berlinale Forum, The Forbidden Room takes as its starting point the reimagining of a smattering of lost films from what one might call the childhood of cinema, encases them in an impossibly complicated concentric structure, and implodes what we usually regard as possible or acceptable in a cinematic “feature-length night’s entertainment.” With few points of comparison, it stands as a proposition for what cinema is right now and what it can be, a fully digital work that resembles some (mostly) imaginary past as processed through the human mind and machine software—like sands in an hourglass, these are the days of our lives.

Ever the accomplished short filmmaker—The Heart of the World (2000) will no doubt make an appearance on the forthcoming list of Canada’s best films of all time, and one of the excerpts in The Forbidden Room explicitly recalls his first short, The Dead Father (1985)—Maddin is best served in feature form when attempting what we might call a digressionary cinema. This traces back to his debut feature, Tales from the Gimli Hospital (1988), through to his 2003 re-envisioning of The Hands of Orlac (1924), Cowards Bend the Knee (to which The Forbidden Room, in its non-stop energy, bears the most resemblance), and even to My Winnipeg (2007), where Maddin created a somnolent, close-enough-to-the truth city symphony by way of side-splitting vignettes. In The Forbidden Room—named for Allan Dwan’s 1914 Lon Chaney-starring short—it’s tied to what can only be considered a perfect melding of subject matter and filmmaker: the reimagining of a handful of lost films, oftentimes only on the basis of their titles and a brief plot summary. (Maddin’s previous attempts at remaking lost Abel Gance films, Odilon Redon [1995] and The Heart of the World, were failures inasmuch as the two titles being remade, La roue [1923] and La fin du monde [1931], actually still exist.)

Maddin is also a filmmaker who owes more than is acknowledged to his partners in crime. In The Forbidden Room, Maddin is reinvigorated by his collaboration with co-director Evan Johnson, a former student whose only previous employment consisted of capping carpet-cleaning fluid bottles. Johnson was also one of the film’s co-writers, and supervised the exhaustive post-production, which still continues: shot at the Centre Pompidou and Montréal’s Phi Centre, the material began as the basis for an internet project, first called Hauntings and now known as Seances, which will appear online later this year. (The impact of the internet on the film’s structure deserves more than just this passing note.) Together, Maddin and Johnson have crafted a formal masterwork jolted by digital after effects, recreating the look of decaying nitrate stock, shape-shifting the image with multiple superimpositions and variegated colour fields (the general look resembling decayed two-strip Technicolor), and compositing swirling transitions that connect (or bury) one film within the other (and the other, and the other).

To try and describe “what happens” in The Forbidden Room is both forbidding and beside the point, for the 130-minute film stands more as an interminable, (in)completed object on its own, like the work of one of its main influences, the French poet, novelist and playwright Raymond Roussel (from whom Maddin and Johnson borrow their technique of parenthetical asides); one comes to understand this object, and what it’s trying to accomplish, only while watching it. (Besides, most of the gags are too good to reveal, but you’ll get one: SQUID THEFT!) Maddin is again in hyper-melodramatic mode here, and each of the all-star cast of art-house fixtures (including Mathieu Amalric, Charlotte Rampling, Ariane Lebed, Jacques Nolot, Maria de Medeiros, Udo Kier, Geraldine Chaplin, and Roy Dupuis) gamely play his or her bit part(s) for shits and giggles, while the film is held together through the multiple appearances of Maddin mainstay Louis Negin (whose John Ashbery-written monologue “How to Take a Bath” frames the proceedings) and newcomer Clara Furey as Margot, the damsel in distress.

Lest one think this is all simply a romp in the park, one of the many salient points Maddin and Johnson make is connecting the mutability of the image (that even extends to the title cards!) to the possibility of endlessly remaking the self through performance; no coincidence, then, that amnesia and memory infuse the film like the smell of vinegar on a decaying old print. Things generally do not go well for Maddin’s perpetually unlucky protagonists, whose troubles are often traced back to childhood, and one could even say that the interconnected chambers of cinematic pleasure in The Forbidden Room are enactments of various personal interpretations of Hell. Feel free to ignore this interpretation, as Maddin and Johnson have ultimately created a supremely hopeful and energetic dreamwork about a cinema that, while it may at times seem in danger of vanishing, will never end—an infinite loop of cinephilia that allows us to lose ourselves in the power and the glory of the big screen and takes us back to our own childhoods, when unencumbered pleasure was easier to come by, before everything got fucked up. As noted in The Forbidden Room, “Nothing is ever the past,” but the past remains a foreign country, and, yes, they still do things differently there.

Cinema Scope: The first thing that I noticed when you had finished this film was the running time, and I thought to myself, “That sounds a bit long.” Then I watched the film and realized, yes, this is long, and exhausting, and this is obviously something you were after when you were editing the film.

Guy Maddin: Thank you.

Scope: It’s great, yet it quite easily could have existed in a shorter version. You obviously had tons of material that you had to condense to a certain extent…

Maddin: You’re right, we could have easily had a 75-minute version. We really battled with that, because there was a lot of pressure after an initial screening which was this length, in September, to make it shorter, and I’ve always desired to make my previous films shorter than they actually are. A director’s cut would be screw-tightening and massive slashing, but this was a different matter, somehow. If we had two instead of three acts, with this Russian-doll concentricity of nested narratives, it would have come in an hour shorter. But then it wouldn’t have been too much, it wouldn’t have been exhausting, it wouldn’t have left viewers, especially viewers that like it—as viewers that dislike it, they’d leave anyway—but viewers that like it, we wanted to feel like we’d broken their brains, really left a physical impression on them, left them exhausted. Hopefully exhilarated and exhausted, in a good way. We wanted “too much” to still be insufficient.

Evan Johnson: Maybe to induce in the audience the fear that it would never end? At least the first time that they see it.

Maddin: There are not many indications.

Johnson: Things are always ending.

Scope: Well, in the Udo Kier father scene we see, what, four endings, leading up to the final, final, final, final ending.

Johnson: There are a series of endings that if you’re confused enough while watching, you could think that it’s the actual ending of the movie but it isn’t. Someone was telling me at a screening of The Turin Horse (2011), which is broken up day by day…

Maddin: With the fades to black…

Johnson: Yeah, a particularly beleaguered audience, after day three, reacted with light applause, thinking that the movie was over…

Maddin: Day three seems about right to me…

Johnson: But then another day would happen, and then another day, and yet another day. I always liked that story.

Maddin: At least we don’t have any slow fades to black…

Scope: There are cuts to black.

Maddin: There are, and actually in that Udo Kier moustache story there are a few slow fades to black, and it’s near the end, but really nowhere near the end.

Scope: So ideally you wouldn’t end the film, is what you’re saying.

Maddin: Yeah, it would be nice if it came out in one endless ribbon, that, like John Ashbery’s poetry, you just snip off for a beginning and an end, and just ask the audience how much they want.

Scope: The other thing about the structure—and I saw the film twice—is that after a while, even the second time, you forget where you are. Again, for you, the film is all about amnesia and memory and forgetting, and in a way this is what the viewer goes through while watching it.

Maddin: I like the feeling when you’re coming out of nested narratives of being reminded where you are. It’s pleasing and rewarding to me. And I only have so many experiences of it. About ten years ago I watched John Brahm’s The Locket (1946), which is about a woman who is a pathological liar, and we see her only through the flashbacks of three different men—Brian Aherne, Gene Raymond, and Robert Mitchum—as they tell their stories to each other. So there’s a man’s story within a man’s story within a man’s story, and then at the very centre of that is a story that she tells to one of the men about a little vaginal locket that is full of the mystery of all women. And I just loved watching it because you really get so caught up in each story that then when you come out of it again you remember, oh yeah, I was two or three narratives deep. It’s kind of fun to get your bearings again, and then get lost again and then remember. There’s the old reliable lawnmower engine of amnesia and memory going on while watching it, and that’s, to me, rewarding to construct. There’s one that I never get tired of in our own movie, perhaps because the gap is the longest of all the framing devices. It’s when that man with really good hearing on that little yellow-brick-road-like journey listens to the snow and starts telling a story…

Scope: And then like 25 minutes later…

Johnson: More like 50!

Maddin: Well, 45 minutes later, there’s a kind of climax where Louis Negin and Marie Brassard are circling each other each with a jar of water to capture Clara Furey’s soul and then there’s a circus rim shot, and then the listening man goes…

Johnson: “Such is the tale of the snow.”

Maddin: And you go, oh yeah, 45 minutes ago this guy started telling this stupid story and we just had to endure eight nested narratives! I still enjoy that one; it just works better than all the others to me.

Scope: That was the story I was specifically referring to in terms of forgetting where the narrative has taken you.

Maddin: Maybe it’s like that Nabokov character in Despair who can only get sexual pleasure by watching himself have sex, so he has to remove himself from his own body, but the pleasure is heightened if he’s watching himself in a mirror, around a corner. The further away…So in a corridor he has to turn a door at a certain angle to see his reflection from as far away as possible…

Johnson: And eventually he gets so far away that he produces his own double.

Maddin: That’s right. So it reminds me of that, the further away in time the two, the gap, the more pleasing somehow. Or maybe I just like rim shots.

Scope: To go back to the beginning, did you always have a feature film in mind for this project? Seances started first as a multimedia project. When I see this film, it seems like you can tell which are the short films that were the products of the filming in the Pomipidou for Seances, but the Roy Dupuis stuff, shot at the Phi Centre, comes across as a framing device that was added to make a feature.

Maddin: I always suspected that something long, like a feature-length night’s entertainment, would be required. I thought it would take the form of the interactive, whatever that might be, of me presenting seances with lost cinema in a live situation as a medium, or some sort of circus clown. But I never really thought that a feature would be necessary. But you’re right, we figured that was the best way. We knew before shooting Roy Dupuis that we needed to make a feature. So we wanted to have a real nice big body of work there. We usually shot one lost film a day, but we spent two days on the Roy Dupuis film, and we made sure that Clara Furey was in three or four stories, to ensure that there was some continuity of character there…

Scope: More like “continuity of actor.”

Maddin: Yeah, exactly. But we were still debating the structure long after shooting in Montréal. We weren’t so optimistic that a feature would be watchable, frankly, but we were just determined to do it.

Johnson: Are you optimistic now that it’s watchable?

Maddin: I’m not sure. I know I can’t watch it anymore, that’s for sure. I know a bunch of polite Germans saw it in a room last night, and most of them stayed.

Scope: As opposed to Sundancians.

Maddin: The DCP was flawed in Sundance, so the sound was a headache-inducing drone. Plus, sure, the Sundance context was different. We wanted to get it at some major festivals, and I didn’t want to wait to get it out, and a lot of American press and distributors go there. I was up for it. And I was prepared for it not being a “Sundance film,” too bad. “It’s not a Sundance film, there’s no hiking in it.”

Scope: There’s lots of hiking in it!

Johnson: The walkouts were exaggerated by the press. Slightly.

Scope: I guess the instance of walkouts in themselves at Sundance was surprising, because nobody really walks out there, they just love anything that’s hurled their way.

Maddin: Well, they walked out on ours. One reporter from the National Post started this story, and wasn’t even at the screening or watched the film! Then I did interviews with people who said, “I hear more than half the audience walked out in the first hour”…the usual viral snowballing was happening. So whatever. I just didn’t feel like talking about it in interviews, it makes you so defensive. But I’m calmer now. Am I supposed to be ashamed of walkouts? They happen, and Tales from the Gimli Hospital and Archangel (2005) had 85 percent walkouts. I know what they mean now, so I’m not concerned.

Scope: The Forbidden Room is in fact the name of a lost Allan Dwan film from 1914. Are these actually all lost films that you’re remaking?

Maddin: To my knowledge they are. Some of them got found when we were making them.

Johnson: In almost every case, the kernel of an idea came from a lost film. Some of them we didn’t know the plot. The Strength of a Moustache, for example, was a 1931 Naruse title, and we invented a story.

Maddin: Yeah, we knew enough about Japanese culture that in Naruse films it was common for a son to feel a lot of shame about his father, so it had to be a father-son story. But then it turned out to be something else.

Johnson: We didn’t cheat that much.

Scope: I ask because in Kino Lorber’s press release, announcing the US acquisition, they write, “Allegedly based on lost films…”

Johnson: Oh, that’s good.

Maddin: That’s our own press release?! Our distributor is busting us. I won’t claim that we did so much research, or that we found the scripts or anything like that. Most of the time we just had the title, for example, Ladies, Be Careful of Your Sleeves, which is also a Naruse film from 1932.

Scope: I think it was about five years ago that a friend was trying to think of museum-based projects involving performance, and I suggested that you could remake lost Méliès films, because there are dozens of crazy, evocative titles: Conjurer Making Ten Hats in Sixty Seconds, A Badly Managed Hotel, The Balloonist’s Mishap….

Johnson: That’s true. We started a script for one, The Shaming of Alfred Dreyfus, and we made the Dreyfus Affair into a time-travel movie. And it was kind of offensive…

Maddin: It was offensive because the Dreyfus Affair is all about anti-Semitism of course, and if you try and make light of it, well…We made the Hogan’s Heroes version.

Johnson: Dreyfus was guilty in our version. That was the problematic part.

Maddin: Yeah, and the time travel, it involved his mother going back in time repeatedly and changing his story a little bit until eventually he was off the hook.

Scope: So what are some of the lost films that are being remade here? I surmise that the story with the aswang is Dalagang Bukid (1919), the first Filipino film ever made.

Johnson: Well, The Red Wolves is the most lost of all…Joseph Roth, the writer, reports in his journalism of having seen this movie in the harbour at Marseilles. Sailors could see a movie screen from their ships in the harbour. So he was talking about this phenomenon, and a movie he called The Red Wolves about forest bandits, from which we took the plot, and we looked and looked, and we couldn’t find anything about it. Maybe he made it up entirely, maybe it’s out there somewhere. We really don’t know. It’s hard to confirm that a film is lost.

Maddin: It’s not like a dead body that you can identify in a morgue. You can prove that it once existed, but you can’t prove that it’s lost…Let’s see, the one black-and-white section is Der Januskopf (1920), a lost Murnau film. Right before he did Dracula as Nosferatu (1922), he did Dr. Jekyll and Mr. Hyde as Der Januskopf. Then it got lost, along with eight others. I think at one point there were 12 lost Murnau films, but then four got found. In Paris we shot a movie that didn’t make it into this cut, a lost Three Stooges film called Hello, Pop! (1933), where they get lost under a giant skirt or something. And it was found in New Zealand while we were shooting it. We shot it as an all-girl production, as we tried to make the movies age-blind, gender-blind, and colour-blind. As it turns out, The Forbidden Room is almost an all-white movie, but the colour-blindedness of the project comes into play more in the interactive version.

Scope: How many films did you shoot?

Maddin: We shot 31 days, one film per day. Some films are mash-ups of more than one film, because there are so many titles that we found intriguing that we wanted to work in there. We were going to shoot 100 days, with a final stretch in New York at the Museum of Modern Art, but then it collapsed, and thank God, because it would have killed me. The one we were both very excited about shooting was called Never the Twain (1974), which was a Brad Grinter exploitation movie about a guy who was possessed by the spirit of Mark Twain while visiting the 1974 Miss Nude World pageant. All that survives is the poster, look it up. And we were going to mash it up with this lost Euripides play where a bunch of women slaughter all the men on an island. And shoot it with the camera Michael Snow built for La région centrale (1971), because I contacted him and asked him if he had any lost movies that I could adapt, and he said, well, I once lost 20 minutes of La région centrale. So I asked him if we could use his camera to shoot this Miss Nude World Pageant slaughter of men.

Scope: The look of the film is something I wanted to get into, in terms of what is digital, and what is in camera.

Maddin: I shot everything on a digital camera. The raw video looked hideous. I really did not have my mojo when shooting.

Scope: You were operating the camera?

Maddin: Me and the DOP. I like to look through the camera while I’m directing, and I like to be able to react on the fly, and think, jeez, I need a close-up here. If you’re relying on just one other person you get mad, why didn’t they push in at this point…believe me, I get mad looking at my own footage too. So with two cameras going, you’re mostly covered. In my filmmaking career I was more comfortable, I quite often had a light leak in my camera, and I refused to clean the lens. There’s a natural light and breathing to film, there’s just some atmosphere in it, and I was getting pretty familiar with how the emulsions reacted to light; I even literally could see in black and white. But looking in the viewfinder when shooting this and seeing some butt-ugly video, it was hard to get my attitude up. And not having any confidence that it could be turned into anything. I finally just shot anyway, with increasing dispiritedness. But Evan kept assuring me that he could probably do…something. And with a lot of post-production labour, he did transform the stuff. He and his brother, Galen, the production designer and the graphic designer, who did the intertitles and also designed much of the soundscape, they worked out some things he can talk about.

Johnson: Well, there were some in-camera effects, we used rear-screen projection a lot, and also some green screen—which isn’t in camera. But there’s no one secret to how we made it look that way…We did research into the history of Technicolor, where it goes wrong technically, so we could learn how colours fade, and are absent…

Maddin: It’s more the colours that you remove that are essential to the look…

Johnson: Sure, because we are working off the look of two-strip Technicolor. I guess our one trick was all of the texturing and colouring, which we did to all of the rushes.

Maddin: That would be like Pixar creating from scratch a 10:1 shooting ratio. They did a lot of hard work just to time 40 hours’ worth of stuff, including the digital slates, which are full of decasia and buckling…

Johnson: We were fairly principled. There were moments that lasted maybe six or seven minutes where Guy left the camera rolling, walked around, filmed people eating lunch, and we graded that too, on principle. Because it felt like cheating to fake film texture like that, so we needed some way of reversing the cheating by making it really difficult again. I was uncomfortable on an ethical level, but once we started to introduce digital degradation, we felt like we were blending the worlds properly and creating a continuity that ought to be there.

Maddin: Yeah, you weren’t just aping something, you were embracing digital artifacting and creating film artifacts. It feels good.

Johnson: You’ve often talked about that year in cinema, 1927, when talkies started to arrive, that in-between period. There was a proper transition between silent film and talking pictures, and it feels like the transition between analogue film and digital has been confused, not in an exciting way. I guess there was agony back then as well. We wanted to aestheticize this transitional moment, I guess.

Scope: Can you go into more detail as to how you created the digital decasia?

Johnson: I won’t tell you how I did it, I really won’t. I discovered it by accident, and I’d never seen it done. The technique is similar to data moshing, which people know how to do, by removing frames, but it’s annoying…but yeah, it’s an Adobe After Effects series of techniques that I accidentally put together.

Scope: Isn’t there one section of the Mathieu Amalric part where the effect is sustained throughout?

Johnson: We wanted to do the whole movie that way.

Maddin: I was just happy to make a two-hour movie that looked like a lava lamp of old emulsions, swirling around.

Scope: You mentioned Nabokov, and to hop back to literature, there are two other authors that came to mind: Raymond Roussel and Franz Kafka. About two years ago I asked if you’d ever considered doing an adaptation of Roussel, as over winter holidays I thought it was a good idea to read Locus Solus, and you said, “Don’t tell anyone, that’s what I’m doing now.”

Johnson: Kafka’s one of the most commonly cited writers when we’re sitting around writing and conceiving. Mainly because he has that beautiful, tiny story…

Maddin: “Give It Up!”

Johnson: Yes, “Give It Up!”, where a man’s lost and he asks a policeman for directions, and the policeman just responds, “Give it up!” That became a mantra for us when we had to abandon something that wasn’t working.

Maddin: There’s so much joy in quitting. Correct me if I’m wrong, but we didn’t think of Kafka so much when we were writing—though there are some stories I think about a lot, like “The Village Schoolmaster,” because I have my own manias, so if someone else comes close to me with a similar mania I get really jealous and petty and vindictive.

Scope: Kafka came to mind because of the mutability of identity that seems to be going on in the film, especially the fact that you have all of these actors playing different parts, and even within the stories these identity questions keep popping up: once you have no memory, who are you? Is your identity a function solely of performance? I thought of “A Report to an Academy,” even.

Maddin: Actually, one of the stories we wrote was for Aamir Khan, which we never shot…He turned us down, what do you know! He got me really excited, then I sent him a script about a guy who was raised by deaf-mute parents on a desert island, so the only thing he could speak were nature sounds, like babbling brooks, bird-wing flutters, but then he was rescued and moved to the big city. This is Ten Scars Make a Man (1924), the lost serial about a guy who gets hit by lightning seven times. And we combined it with…what is the name of that first drug casualty in Hollywood history? There was this guy in 1923 who died of an overdose, he was in a lot of racing-car movies. [Ed. note: Wallace Reid.] Anyhow, this guy who could only speak in nature sounds becomes a champion race-car driver. That’s the closest to “A Report to an Academy” that we came.

Scope: And Roussel?

Maddin: Roussel was of course never far…maybe it traces back to The Locket. After watching it, I remember thinking, geez, if I were a filmmaker, I would try and make something with stories within stories within stories, and then I remembered, well, I am a filmmaker, I’m gonna do that one of these days. And that put me in mind of Roussel, and also John Ashbery, who I had just been reintroduced to, and so I wanted the internet interactive version to be kind of Rousselian. I wasn’t really sure what that meant; Evan had read far more Roussel and understood the structure better. But very early on in my friendship with Ashbery, I invited him to write a script for a lost movie. I gave him a long list of titles, and he chose Dwain Esper’s How to Take a Bath from 1937, made by the same guy who shot How to Undress in Front of Your Husband, which still exists. How to Take a Bath supposedly just compared, in split screen, two women, one of them married and therefore sexually experienced, and how she took a bath, with an unmarried woman, who’s sort of just one big hymen, who is giving herself an uncomfortable soak. So I described that to Ashbery, and he wrote this monologue. I shot it with Louis Negin when I was shooting Keyhole, back in 2010.

Scope: It does look different to me.

Maddin: Yeah, it was shot with a different camera, and I guess colour-timed slightly differently…and Louis was in his callow youth back then. So it seemed to tie in because Ashbery introduced Roussel to the English-speaking world. When he moved to Paris as a 27-year-old he discovered the existence of Roussel but he couldn’t read him, because he hadn’t been translated. So he learned French to read Roussel, then translated him, and he became a lifelong passion. But it just seemed to make sense that of all the parenthetic embraces in Roussel’s New Impressions of Africa, and all the parenthetical embraces in our movie, that the outermost parentheses would be John Ashbery’s “How to Take a Bath.”

Johnson: What I liked about him always, my favourite thing of his, I don’t know really what it is, is Documents in Service of an Outline.

Maddin: Oh, it’s exquisite.

Johnson: It’s an incomplete novel, but it has the most intricate structure, some stories get ten or 11 layers deep…

Maddin: And some stories are just one sentence long!

Johnson: It’s a play within a dream within a tapestry. He put an enormous amount of work into these narratives, and you’d think that at the centre there would be some kind of payoff, but there rarely was, just an emptiness…most people would say that’s a flaw, but it’s unnerving. Roussel died holding a key in front of a locked door, and nobody knew whether he had locked himself in or was trying to get out before he died. So our structure was partaking of the mystery of something with no core—that’s why it’s hard to decide if the film is about anything, or just about it’s own shape.

Maddin: Yeah, we decided at the centre of the third act would be Charlotte Rampling because she was the mother of the Ostler in that act. But our editor John Gurdebecke lobbied hard to make her the chewy centre of each concentricity, and we thought it might be a bit of a red herring, frankly. Not that I’m against red herrings…

Scope: Instead you have this pursuit narrative of Roy Dupuis saving Clara Furey, with all of these narrative obstacles to traverse.

Johnson: And when they embrace at the end, that feels empty to me.

Scope: The very end is completely anticlimactic. But of course before that you have the Book of Climaxes.

Johnson: That’s why I like that. It renders the guy getting the girl as the limp cliché that it is.

Scope: So you need to have the guy getting the girl while two dirigibles smash against each other in the foreground—now that’s a climax! I take it those are scenes from other films that you shot that didn’t make it into the film.

Maddin: Some of them.

Johnson: Others were just climaxes: we sat around one day writing what seemed like climactic material, and then shot it all in one day.

Maddin: Evan suggested, and I reminded him it would likely be in contravention of Telefilm Canada contracts, that we could just make another feature with some of the other material, and instead of having a Book of Climaxes, have a Book of Cold Opens at the beginning, then a Book of Tipping Points, or a Book of McGuffins. There are so many things straight from the Robert McKee text that we could pile up like signposts.

Scope: What was it like directing in front of a live audience?

Maddin: At first I was very proud, because I was in the Centre Pompidou, which has a fine pedigree, though we were next to the cruisiest bathroom in all of Europe. Shooting in front of an audience was one thing, then running to the bathroom with Mathieu Amalric and having to hack your way through the marshland mating rituals and the heroin abuse was always interesting. I learned bladder control. After a while it was like we were contestants on some kind of reality show: you just forget that the observers, whether they were cameras or people, were there. Every so often I’d look out and see someone I knew, someone passing through Paris from Winnipeg…There were a couple of alarming psychopaths who were there every day…

Johnson: Fans.

Maddin: Yes, fans of mine. You know when I first picked up a camera, I dreamt of having fans.

Scope: Did you shoot with sound?

Maddin: We loved it! It was so noisy in there because there are thousands of people in the foyer, and their voices combine into a seashell mush, it’s like hearing the ocean. On the first day of shooting it was the first day of a “festival,” and a radio program was hosted right above us, and they were playing Dire Straits, Sting…and so there’s dialogue with “Sultans of Swing” in the background. Our editor managed to remove much of that, but it’s there all right. A real sound editor would go berserk. Dreyer shot Vampyr (1932) in three different languages, and in the English version the dialogue was just sliced into the music, so the music would go right up to the very frame where the voice started. So the music would suddenly drop out, then you’d hear, “There are no dogs here!” Then brrrrr, it would begin again. We made a point of doing that in the submarine sequence. It’s a real thrill to have cut-and-paste sound editing. I love Oscar Michaux sound editing, he’s the best of them all—none of this Walter Murch crap!

Scope: What were you after in terms of the music and the original composition?

Maddin: I did a lot of loops. It’s a matter of taking public-domain recordings, finding a section you like, and then endlessly looping it. I guess the one I’m most proud of plays around the two-hour mark, when Mathieu Amalric is changing clothes with Udo Kier in real time—it takes about 45 minutes—there’s a short loop from an old opera aria from Ezio Pinza, with some fakey wind effect on top of it. But Galen wrote most of it in far more sophisticated, detailed mixes of atmospheric things.

Johnson: It was the aural equivalent of the visual composition. Guy would take an old version of a Brahms piece and stretch it out, but he was often using a source file that was low kilobytes, and you can hear in the movie all the digital degradation to the piece. But it sounds kind of nice, if you get used to it.

Scope: What form will the internet version of the project take?

Maddin: Seances will be launched around the same time as the film is released. It’ll be a website where anyone visiting can hold a seance with lost cinema. The lights will go down on your screen, and then you’ll see the start of a fragment of one of these lost films. Just like in the seances we all know about from movies, the strength of the signal of one spirit might fade while another collides with it and takes over, so there’ll be a few non-sequitor collisions or collages. We’ve done some tests, when you have two or three stories with some brief interruptions from other stories coming in, it creates an impressionistic narrative after a short time, 15 minutes at the most…which is a long time for the internet. But too bad!

Scope: The interesting thing about the whole project is that the jumping around even within the film, from story to story, has this internet-narrative quality to it.

Maddin: Yes, and we were even toying with the idea of having so-called “weak spots” within the film: at one point if you touched your iPad, you could push from one story to the next, or there were even weak spots with YouTube clips or weather reports. That may still happen, we’re still fine-tuning. It’d be nice to have a cat video from 2007 in the middle of one of these things. Anyhow, there are like 500 billion permutations even without the cat video, so each time you visit you see a unique combination, and then the program destroys that combination. So it creates a movie, names it actually, there’s a metadata thing—the first name it came up with was Wise Trumpets of the Milky Midnight (2014), which has been logged and destroyed already, so it’s a lost film. So the program creates a film and then loses it, all in one sitting.

Scope: Lastly, can you elaborate upon the impact that digital editing has had on your filmmaking? Could this have been done the old-fashioned way? I’m of the opinion that it has a negative impact on some filmmakers, but for you it seems to have invigorated things.

Maddin: Yes. The Heart of the World was the last thing cut the old-fashioned way, with work prints, china marker and tape. I did it on my porch with deco dawson. So much fun, and there was so much dust and fibre crap, not to mention what looked like eyebrow hair, floating around in that dingy chamber, more like the chicken coop in which Link Wray recorded his albums, and all of that airborne crap ended up embedded in the negative because deco rephotographed the work print on an optical printer, thus making the movie impossible to “restore,” and also saving us a big bill from the negative cutter.

But since 2000 I’ve been working in digital. At first I didn’t exploit its potential much. We slowed down some of the dancers’ pirouettes in our Dracula ballet work, but that was just something an old step-printer could have done. The real exploitation came with Cowards Bend the Knee, when my new editor John Gurdebeke and I discovered we actually liked the look of digital video when one is searching in fast forward, or rapid reverse, for a specific shot. We liked the skimming over the images, the way Final Cut omitted much of the footage while in that hurried mode, much like a skipped stone elides water; we were experiencing ellipses galore during the cataloguing of rushes. Since I’d been living for years in an apartment plastered in Post-It note reminders to “BE MORE ELLIPTICAL,” this artifact of Final Cut thrilled me.

I also fell in love with the way the picture slowed down to fetishize an object once we started zeroing in on its location. We often rushed past it in our haste, so we had to stop and back up, then when we went past it going the other direction there was another halt and U-turn, before we finally found it and slowed the picture down to study it. This whole process reminded me of the way one revisits a truly precious memory, at first in haste, then again with more methodical determination, then finally and repeatedly with a long fetishizing languor. The act of simply searching for a shot became the syntax for sliding back into memory in Cowards. This could have been done the old analogue way, but it would have been like me trying to draw every cel in Pinocchio (1940). Final Cut made it easy. The footage we cut together was really repurposed rushes, and this stuff was produced in a process more like VJing a happening than like shooting. The shooting was ordinary, but the recording of our improvised playbacks gave Cowards its insane elliptical energy. It’s maybe the work of which I was proudest until The Forbidden Room came along.

Now I have the confidence that a lot of effects can be achieved in post, so I just let my caméra-stylo run wild on set. I’m basically just sucking up images, the way someone armed with an old ’90s-household Dirt Devil might suck up dust bunnies from the floor of the set. And I do so with the confidence Evan or Galen will be able to make the hideous raw video look exactly the way we like it at the post-production digital-effects stage, and Gurdebeke will be able to do his ersatz-memory editing magic. As a shooter I’m completely liberated to improvise, finding my shots as I go, knowing I’ll have a million digital options to fix all my mistakes, or even convert them into coups. I could never have done any of my 21st-century work the analogue way. Having said that, a vast majority of my life-altering film-viewing experiences were made the old way. But what am I supposed to do? I don’t have Josef von Sternberg’s talent. All I can do is scratch and claw with every little advantage the new technology gives me.

Mark Peranson