A Russian Filmmaker Proved the Audience Acts the Movie

Around 1918, in a Moscow editing room, the Russian filmmaker Lev Kuleshov spliced together a short demonstration. He used an old close-up of the matinee idol Ivan Mosjoukine — a blank, neutral face — and cut between it and three other shots in turn: a steaming bowl of soup, a girl lying in a coffin, a woman on a divan. The audience watched the same Mosjoukine close-up each time. They reported that he was clearly hungry, then grieving, then desirous. They praised his subtlety.

The footage of Mosjoukine never changed. Kuleshov had spliced the identical strip into all three sequences. What had changed was the cut — the meaning the audience built between the two shots. The viewers were doing the acting and giving Mosjoukine the credit. Kuleshov took it as the basis for what he believed cinema actually was: not a recording of performances but a controlled construction of meaning by editing.

His students at the Moscow Film School pushed the idea into doctrine. Sergei Eisenstein turned it into the theory of dialectical montage in Battleship Potemkin (1925); Vsevolod Pudovkin and Dziga Vertov built different schools of editing around the same insight. Decades later Alfred Hitchcock liked to demonstrate the same trick to François Truffaut: cut his face to a baby and he is a kind old man; cut it to a bikini and he is a dirty one.

The original strip is gone. But every reconstruction — and at least two psychology studies, an fMRI experiment by Dean Mobbs in 2006 and a behavioral one by Daniel Barratt in 2016 — has produced the same result. Viewers really do see a different face. The face is not different. The audience is.