Tempo rubato is one of a number of techniques used by musicians to accentuate an important moment in a passage of music. Film, too, has many devices for accentuating important moments -- such as close-ups, sound effects, etc. -- but rubato either isn't one of them, or it hasn't been much used by movie-makers to date. This omission is a shame, since the technique would seem to transpose very well from the one medium to the other. (It may be in fact that many directors have already made extensive use of this technique; I'm not a film buff, and I've only made a cursory look for it.) The reason I'm writing this essay then is because, whether or not the technique is already known in the film world, it's worth spreading the word about it still further.
The term itself is Italian for "stolen time", and in a nutshell, it accentuates a moment in music by speeding up the tempo ("stealing" time from the initial pulse or tempo), then slowing down ("giving back" the time), before returning to the original tempo right at the critical moment (usually either the climax of the passage, or the ending). The effect is similar to riding a bike along a level surface, then going up and down a hill (or down and up a valley, which is the more common method in music), before returning to level ground again; except in music it usually happens much faster than this. It's the somewhat lurching (up and down), or turbulent effect, followed by clear sailing (level ground), that makes it work. (The technique is probably most easily heard in the music of Chopin, and anyone interested in the detailed mechanics of it should read Tobias Matthay's Musical Interpretation.) The salient points apropos this essay are: rubato needs to be so discrete as to be nearly subliminal, and, it must involve both speeding up past the original tempo, and slowing down slower than that original tempo.
In case some might raise an eyebrow at the idea of transplanting an idea from one art to another, it's best to remember that the arts have always cross-fertilized, on account of the many parallels among them. Architects have always taken ideas from painters and vice-versa, musicians and dancers have done the same, etc. Music and film also make a good couple -- as they both feature the element of time -- and so it's not hard to make a long list of techniques and ideas that may be freely transposed from one to the other: the film techniques of cross-cutting, long takes, and even things like racking focus easily find their parallels with music in the mind of an inventive composer. (No doubt this is why film composers work carefully with screenwriters and directors.) And in fact many composers have been directly influenced by film: Elliot Carter, for example, was influenced by Sergei Eisenstein, and no doubt many in movie-making have been directly influenced by musicians.
Slow motion, as an ancestor of the technique in question, has been in use for some time: Kurosawa famously manipulated the visual tempo in the duel scene towards the start of Seven Samurai. I've also noticed that the editors of videos of surfing competitions seem to have an intuitive leaning to this technique, since on the replays the athletes often return to real time from slow motion at the critical moment. But they and others leave out the "stolen" part, i.e. they don't speed up before slowing down (or vice-versa), and it's this "back and forth" that takes really makes it work.
To create a hypothetical example of how rubato in film might work: imagine a joust, beginning with both riders charging at each other from a long distance. As the moment of impact continues to approach, the tempo is speeded up gradually (but only slightly overall), until the riders are getting still nearer to one another, at which point the tempo slows down -- also gradually down to only slightly slower than real time -- only to return to strict tempo as the lances crash into the shields and the combatants go sprawling.
Of course, cross-cutting, camera proxemics, etc., would all play their roles here, just as dynamics -- the loudness of the passage -- harmonies, etc., all play their roles in musical rubato: they all work together to heighten the aesthetic effect. In fact, a more sophisticated version of rubato would involve a film composer writing a rubato passage to go with the rubato on screen, both composer and screen writer using all their devices to really pack a punch.
But just as some styles of music take better to rubato than others, so some styles of film-making will also do better than others with the technique. Action scenes -- whether dramatic or comic -- in particular would likely take well to the technique: everything from high-speed chases to sports. Romantic scenes, on the other hand, would probably wither under it, as would dialogue. It seems the main criterion for its success is whether or not the scene in question involves time in a some crucial way.
Perhaps the reason it hasn't caught on is that some believe that rubato in film would be too artificial, that it would "activate disbelief" in the viewer. This seems a flimsy objection, since movies have always featured plenty of artificial effects that work very well, everything from using black and white when color is available to cgi's of dinosaurs. All of those things -- when done well -- tend only to enhance the quality of the film.
In music, time is "stolen" from the pulse, or beat, of the music, either while that pulse is sounding (a melody in tempo rubato played over a bass line in strict tempo) or after it's been established (a passage with all the parts -- melody, bass, etc. -- all in tempo rubato played after a passage with all the parts in strict tempo). Some might on those grounds object that there's no "pulse" in a film scene from which to steal, and that the whole idea should be accordingly discarded. But going back to the joust example: everyone has an idea of what it looks like for a horse to run at top speed, as well as an idea of what real time looks like versus fast- or slow-motion. These ideas, or expectations, provide the "pulse", or regularity, against which the rubato can take place; similar to the way that, when listening to music, everyone accepts the initial pulse and sets their expectations against that for what follows. In other words rubato is a means of playing with expectations in general and with tempo only in particular.
In fact, of the two types of rubato -- those with and without a strict bass accompanying the melody in rubato -- it seems that, in film, it's the latter type that would succeed more often. With special effects, however, it seems possible to speed up the main action (the jousting combatants) while keeping the background (spectators, trees, etc.) in strict time, though the effect is likely to go unnoticed.
Tempo rubato has been tested in music for centuries now, and it's well-worth making use of it in film as well. Perhaps it already happens all the time -- I don't know, and pun intended -- but I hope this essay will help to ensure that it happens even more often.
