La Flor · essays & theory

A reading · through the lens of theory

Before anything else happens, someone sits down with a notebook and draws a flower.

That's Mariano Llinás, the director, and the drawing is the whole plan. Six stories. Four of them start and never finish. One has a beginning and an end. One is nothing but an ending. Sketched out, the shapes make petals. Then the notebook closes and you are handed fourteen hours: a cursed mummy in a provincial laboratory, a pop singer caught in a conspiracy about scorpion venom, five hours of Cold War spycraft, a film shoot coming apart, a wordless black-and-white pastoral, and a nineteenth-century captivity story told through optics so soft that faces dissolve into shape and light.

Most films hide their structure and hope you feel it. This one shows you the structure first, then makes you live inside it.

Notice what the drawing is not. It isn't a plot. Plots are lines. It is a picture of how six blocks of time will be shaped, and its meaning keeps shifting as you travel it, because the petals you've finished stay open behind you. In Deleuze's vocabulary that's a diagram: a plan that maps time instead of space, and that changes as you move through it.

You could describe La Flor without Deleuze. Anthology film, genre pastiche, endurance test. That description survives about twenty minutes, until the first story stops dead near what should be its second act and you have to decide whether you've been cheated. Deleuze is worth bringing because he offers a better answer than "the director is being difficult." His argument, roughly: for its first fifty years cinema ran on a circuit. A character sees a situation, the situation presses, they act, the action changes the situation. He called this the movement-image, and he thought the ending was its most load-bearing and least honest part, the place where a film promises that acting settles things. After the war, filmmakers began declining the promise. La Flor keeps every other component of the machine. The mummy has a curse, the agents have a mission, the venom has a plot behind it. It simply unbolts the resolution.

Which raises the question of what fills five hours of espionage if not the mission. Digression. The spy episode suspends its present tense again and again to give each of four agents a long backstory, and the backstories don't explain the mission, they compete with it. Deleuze separates two ways a film handles the past. There is the recollection-image, the dated flashback: you go fetch a memory, it clarifies something, you come back. And there is what he called sheets of past, where the past stops being a set of retrievable facts and becomes a territory with regions, all coexisting, none subordinate to now. Alejo Moguillansky's cutting builds the second thing. The mission survives as a spine for digression rather than a line to be followed. The craft debt is exact: Raúl Ruiz's Mysteries of Lisbon, where every new character halts the film to deliver a backstory that spawns another, and behind that The Saragossa Manuscript, tales opening inside tales until the frame narrative never gets back to close.

The conspiracies themselves are motors. Rivette had already proved in Out 1 that you could set one running purely to keep actors in motion for thirteen hours, then walk away from it. Deleuze's phrase is the condemnation of the plot: a diffuse plot-behind-the-plot, felt everywhere, resolvable nowhere. La Flor runs two and abandons both without embarrassment.

Now the part that makes the film moving rather than merely clever.

Four women appear in all six episodes as entirely different people, and the change is worked with wigs, costume, posture and accent. Llinás never disguises the apparatus. The spy film looks like a spy film the way a well-mounted stage production looks like the world: by agreement. Deleuze would call each of these women a forger, not a liar but someone whose truth is the power to become someone else, serially, in front of you. The wig is a gest, a posture that shows you the theatrical relation holding it up. You are never fooled and never meant to be. The serial-disguise machinery comes straight from the crime serial, from Les Vampires, where a masked woman is the only continuity between installments that end by withholding resolution on purpose.

There's a stricter contract underneath. Bazin's old claim for photography was that the image bears the trace of what actually stood before the lens: the fiction can lie, the camera can't lie about presence. La Flor splits that contract cleanly in two. Everything the stories say is frankly counterfeit. Everything the camera records is true, including the unhidden fact that shooting took most of a decade and Elisa Carricajo, Valeria Correa, Pilar Gamboa and Laura Paredes visibly age inside the film. Six fictions laid over one documentary of four working lives. That indiscernibility is Deleuze's crystal-image, and the fourth episode makes it literal by filming a film shoot falling apart with the camera behaving like a crew member. The word for what the actresses are doing is fabulation: people caught in the act of legending themselves. Piel de Lava's devised ensemble practice is the method, and behind it stands Céline and Julie Go Boating, where the fiction is written outward from what the performers invent. The women, not the story, are load-bearing.

The voice-over is the engine that made all of this affordable. Llinás worked it out in Historias extraordinarias: a narrator who doesn't explain the image but runs alongside it, supplying the plot the camera declines to dramatize. Sound and image stop illustrating each other; each holds its own. It is also, bluntly, a financing strategy. An unfunded collective can stage an epic by telling you most of it while filming a woman crossing a field. That field matters. The pampa is flat, horizon-dominated, offering no cover and no destination, and when La Flor opens out it opens onto emptiness. Deleuze's any-space-whatever: space that has quit its job as a setting for action and registers as pure feeling instead.

The last episode pushes until the image nearly goes. A nineteenth-century captivity narrative, near-imageless, pinhole-soft, a voice speaking a buried past over a landscape that can no longer show it. Deleuze called this kind of shot stratigraphic, archaeological: an image you read rather than see, carrying layers of time it cannot picture. The film's final story is only an ending, and it looks like a document that survived from before photography properly existed.

Then forty minutes of credits over landscape. Nothing advances. Dead time raised to a principle, a joke about labour, and a refusal to release you on cue.

What it did: it proved that a fourteen-hour film could be financed by refusing financing, shot by a crew small enough to travel, and exhibited in three parts without pretending the parts were separate films. It didn't found a school. It demonstrated a method, and the method is available to anyone with four collaborators and ten years.

One thing to carry into a rewatch. Of six stories, exactly one has both a beginning and an end: the silent pastoral. It's a remake of Renoir. The only whole story in the film is the one Llinás didn't write.