How did Soviet filmmakers turn two clashing shots into an idea?

Soviet montage theory, developed by filmmakers in the 1920s, makes a bold claim: the force of a film lies in the editing of shots, not in the content of any single shot. Montage is French for assembly or editing, and Sergei Eisenstein called it "the nerve of cinema". His version became the most widely accepted: montage is an idea that arises from the collision of independent shots.

Eisenstein felt that this collision could manipulate the emotions of the audience and create film metaphors. In Strike, images of violence inflicted on workers are cut together with images of a cow being slaughtered in an abattoir, and the clash is meant to produce revulsion at the oppression of the workers. In Battleship Potemkin, shots of a sleeping lion statue, a waking lion and one rising to its feet are edited in succession to create the illusion of a statue springing up, symbolizing revolutionary awakening.

His colleague Vsevolod Pudovkin disagreed. For Pudovkin a film is not shot but built from the separate strips of celluloid that are its raw material, and images build upon one another over the course of the film instead of colliding. Where Eisenstein treats the clash as a negation, Pudovkin treats it as an addition.

The difference with the continuity editing of Hollywood is basic. Continuity serves a predetermined narrative, while montage holds that the collision of images creates a film's meaning. This is why, in Soviet theory, time is subordinate to the symbolic meaning of the images, and why the Rocky training sequence, though people call it a montage, is something different.