Clint Bentley and Greg Kwedar are longtime filmmaking partners who have collaborated for over 15 years, making films like Sing Sing, Jockey, and Train Dreams. This conversation covers their creative process from building a mental reference bank of cinema, to adapting a novella, to writing less dialogue, to trusting the process of filmmaking. Their partnership is defined by complementary temperaments: Greg pushes to move forward and get ideas out, while Clint insists on refinement and not settling until the last possible moment.
Building a reference bank through lifelong film consumption
Both filmmakers grew up watching mainstream blockbusters, but their real education began when they started thinking of themselves as practitioners and deliberately sought out classic and overlooked cinema.
Clint introduced Greg to classic cinema; Greg kept a pulse on contemporary film. They shared discoveries and shaped each other’s tastes.
Greg’s personal awakening came from seeing E2 Mama Tomen, which showed him a film could “allow the world to breathe” and be about simple things that are also deeply meaningful. It aligned with his ambition to make films that feel like adventures.
They now seek out pockets of film history they haven’t explored, finding that silent films and films from the 1920s and 30s often take more risks than contemporary work, countering the assumption that art always moves forward.
Staying childlike as an artist matters: Greg cites a Picasso exhibition of casual doodles and constructions made over breakfast, showing how constant experimentation keeps creative avenues open.
Borrowing and adapting ideas from other films
On Train Dreams, cinematographer Adula Velos and Clint solved a framing problem on set by recalling a shot from Tarkovsky’s Ivan’s Childhood where jagged boards frame the boy. They found a curved burnt stick on the ground, placed it between camera and Joel Edgerton, and shot low. The shape made the frame click and visually echoed the character’s grief and entrapment in a burnt forest.
This kind of on-the-fly invention, pulling from a deep well of film knowledge, is central to their process.
The mission statement: “Stories of human connection in impossible places”
Early in their careers, inspired by business school thinking, they wrote a mission statement: to tell stories of human connection in impossible places.
It functions like a personal value system rather than a conscious filter. It doesn’t limit; it orients. It helps them choose projects as opportunities multiply with success.
For Train Dreams, the “impossible place” is a brutal, rapidly industrializing logging world that grinds workers down. The mission statement serves as a reminder to fight cynicism and find light, even if it’s just a glimmer, so the story can endure.
Adapting a book for film: Train Dreams by Denis Johnson
The novella is only 117 pages and deeply beloved. Their rule was to be true to its spirit but let everything else go.
Adapting across media is like switching from walking to swimming: the rules are completely different. What works on the page doesn’t necessarily work on screen.
They started by transcribing sections directly (like the man shot by his dog), but as the narrative developed, those passages no longer fit. You have to listen to what the work wants to be.
They believe you must first deeply investigate the text, then go on your own journey to understand what’s between the words. They traveled to where the story is set, read logging journals and Forest Service ranger accounts, and spent time in the world.
They found a cabin on the river near where Denis Johnson lived, hired a naturalist to teach them about the area’s ecology, and spent nights in logging-town bars talking to locals. One encounter with a Kootenai tribe member working to reintroduce sturgeon led directly to the image of a 10-foot fish hanging from a pole at the beginning of the film and influenced the character of Ignatius Jack.
They listened to the audiobook narrated by Will Patton (who narrates the film) while driving through the landscape, which recontextualized passages they’d read many times.
They finished the first draft there during a snowfall with trains running at night, riding high, then read it on the way home and realized they had a lot of work to do.
The iterative writing and editing process
Writing is like building IKEA bookshelves with custom trim: it looks terrible at first, but you keep refining, molding, painting, adding details until it comes together. The founder of Rolls-Royce reportedly said he just looked at a normal car and made every single thing better, over and over.
Clint is known for not settling, pushing every moment and line until days before a premiere. Greg pushes to get ideas out the door. They balance each other: Greg helps Clint see beyond the draft to how great it could be; Clint helps Greg refine rather than rush.
Greg came from a fiction-writing background focused on the sound of words. Clint taught him that structure and story are foundational; everything else flows from there.
Writing less dialogue
A mentor advised them to use dialogue only as a last resort. Early on this didn’t make sense, but they learned by doing.
In their first film Trans Pecos, a long dialogue scene in a truck became one line and some looks, and it was far better.
Train Dreams embodies this principle: the film is sparse with words, but when dialogue comes (like the preacher-in-the-pulpit line or the Ferris wheel monologue), it lands with force.
The exception that proves the rule: there’s a scene where a character talks continuously, but the point isn’t to follow the words; it’s to feel him spewing energy into the world.
They overwrite in the script so that actors, the DP, and production design internalize the full context, even if it’s ultimately conveyed through a look or an image. You have to write it to know what you’re reaching for.
A proud voiceover about glaciers and deep history was cut because, as their sound designer said, it was good writing but not a good scene. You choose the good scene.
Letting the story guide you
At a certain point the story takes on its own life, and you feel like you’re finding it rather than creating it.
The original ending of Train Dreams followed the book’s wolf boy sequence almost verbatim. They shot it. But in the edit, it became clear the airplane sequence was the actual ending; everything after it felt like 10 extra minutes. They resisted at first, then leaned in, and the film became richer and deeper.
The mirror scene (Grineer seeing himself for the first time in years) was never scripted. Their production designer found a basement bathroom, they shot something, and it opened up new thematic possibilities.
Across their four released films, every single one has a different ending than what they originally shot. Endings are the most important thing; that’s what the audience walks out with.
Calibrating emotion: not overdoing it
A central premise of Train Dreams is treating real life as cinematically as any Hollywood plot: falling in love, a peaceful day with family, the feeling of walking out of a movie that changed you.
The goal is resonance without schmaltziness, depth without manipulation. There is inherent manipulation in any art form, but the aim is not to be dishonest about it.
The film surprised them: they thought it would be interesting but not deeply emotional. Audiences have cried. People who watch it multiple times get something different each time, not because the film changed but because they changed.
On the love story between Gladys and Grineer, they calibrated carefully: showing early infatuation (can’t keep hands off each other) alongside later arguments over money. Too many barbs in early cuts made them seem like they didn’t like each other; too much sweetness didn’t feel real. The edit required finding the balance.
Getting feedback
Early on they were defensive and emotional about notes. Now their approach is: just try it, see what’s there.
They rely on a small think tank: a friend named Linda Halbert who has read almost every draft, plus select collaborators and financiers.
They value two kinds of feedback: from people who know what they’re going for (and can say “this isn’t coming across”) and from people with no context (closest to a real audience).
Sometimes you can just watch someone watch: leaning forward means they’re in, retreating means they’re not.
Feedback shakes the tree. Someone may point at a symptom when the real problem is earlier in the script, but it surfaces where the work needs to go.
Soundtrack and music
Both develop playlists during the writing process from wide-ranging sources. They create mood reels with early dailies or select shots set to music as guiding tone poems.
They use temp music in editing, then pull it all out to give the composer (Bryce Dessner, on all their films) a blank slate, while sharing the guiding principles and early references.
The goal is for music to feel like all of a piece with the film, so the audience can’t tell what was improvised and what was written years earlier.
Mantras they return to
“Pay attention”: Clint’s father’s advice, which applies to writing, editing, production, meetings with actors. Everything can be solved from paying attention.
“Trust the process”: Borrowed from the rehabilitation program at the center of Sing Sing. The process is where the life is. Make it good, healthy, and intentional. Go on adventures into the world you’re writing about. Imbue the work with the values you try to live with.
Focus on details and the bigger picture takes care of itself: You might not know how to pull off a big sequence, but focus on the granular details in the writing, the shot list, the edit, and the sequence will play out.
