How a Falling Cat Turns Mid-Air With Nothing to Push Against
Updated: 10 hours ago
Picture a cat, upside down, held still in mid-air. Now let go.
There's nothing for it to push against. It has no spin to begin with. It's just falling.
And yet, a moment later, it's the right way up, legs underneath it, ready to land.
That should bother you a little. Because there's a rule in physics that seems to forbid exactly this.
The rule the cat appears to break
The rule goes by the name of conservation of angular momentum. In plain words, angular momentum is the amount of spin something has. And the rule says that, unless something outside pushes or twists it, an object can't change its total amount of spin.
You've seen this in action if you've watched an ice skater. Arms out, a slow spin. Arms pulled in, a fast spin. The skater can change how fast they turn, but they can't start spinning from nothing, and they can't stop without something, like the ice, to push against.
So a falling cat that starts with no spin should, by this rule, stay with no spin. It should land on its back.
It doesn't. And for a long time, that was a genuine puzzle.
Caught on camera
In 1894, the French scientist Étienne-Jules Marey decided to look closely. He was a pioneer of something called chronophotography, which is a way of taking a rapid run of photographs, one after another, to break a fast movement into stages. He photographed a falling cat at 12 frames per second.
He published the photographs, and his conclusions, that year in Comptes rendus, the weekly proceedings of the French Academy of Sciences, under a title that translates roughly as "On the movements that certain animals make to land on their feet when dropped from a height". A summary appeared in Nature the same year.
The pictures settled one question straight away. People had suggested that the cat must be cheating: pushing off the hand of whoever dropped it, to give itself a little twist to start with. Marey's photographs showed that the cat had no rotation at the start of its fall. It wasn't using the dropper's hand as a lever. And he concluded that air resistance played no part either.
So the turn really did happen in mid-air, with nothing to push against.
Marey's own explanation was that the cat uses the inertia of its own body, with its front legs and back legs taking turns to act as a kind of pivot. But not every physicist was convinced, and some went on believing the cat was somehow cheating.
The trick: stop being rigid
Here's the key. The no-spin rule is strict about the total. But it doesn't say anything about how the parts of a body move relative to each other.
If the cat were a solid block, it would be stuck. Whatever it started with, it would keep.
But a cat isn't a solid block. It has a very flexible spine. And a body that can change shape can do something a rigid body can't: it can end up facing a different way, while its total spin stays at zero the whole time.
It's a bit like the way you can turn round on a spinning office chair, with your feet off the floor, just by swinging your arms and twisting your body in the right order. You never push against anything. You just rearrange yourself.
Bend, then twist
So what does the cat actually do?
It bends in the middle. Think of the cat as two halves, a front half and a back half, joined by a bendy spine. Once it's bent at the waist, the two halves can each swing round on their own axis, and they move in a way that partly cancels out. When it's all done, the whole cat has ended up the other way up.
At no point does the cat as a whole have any spin. It's never rotating, in the sense of the rule. It has simply changed shape in a clever sequence, and the sequence leaves it facing a new direction.
Shown with two cylinders
Describing that in words is one thing. Showing that it really works, using nothing but the laws of mechanics, took a surprisingly long time.
The physics was first worked out convincingly in 1969, by T. R. Kane and M. P. Scher of Stanford University, in a paper titled "A dynamical explanation of the falling cat phenomenon". The work was supported by a NASA research grant.
Their model of a cat is wonderfully simple. Two identical cylinders, one for the front half and one for the back, joined at a single point. The spine is allowed to bend but not to twist, so the two halves always turn by the same amount as each other.
They then set the total spin of the whole system to zero, because the cat starts at rest. And they kept it at zero throughout.
The model turned over anyway.
There was one condition. In their calculations, the cat couldn't right itself unless it bent backwards a little as well as forwards, and by less. In their worked example, the forward bend was about 116 degrees and the backward bend about 25 degrees, which they treated as roughly the most a real cat's back could manage.
It's a good example of how a deep idea can hide in an everyday sight. You can turn without spinning. You just need to be able to change shape.
Still being studied
That doesn't mean everything about it is settled.
Real cats almost certainly combine more than one trick, and the exact mix is something researchers have kept studying. The Kane and Scher model captures the core of the move, but a real cat is a lot more complicated than two cylinders.
And it's worth saying clearly: this is about the physics of the turn, not about cats coming to no harm. It isn't something anyone should test.
Rearranging the world
So how does a falling cat turn over with nothing to push against?
It doesn't push. It doesn't spin. It bends, it twists its two halves against each other, and it ends up the right way round, all while obeying the rule that seemed to forbid it.
Sources
T. R. Kane and M. P. Scher, "A dynamical explanation of the falling cat phenomenon", International Journal of Solids and Structures 5(7) (July 1969), pp. 663–670. doi:10.1016/0020-7683(69)90086-9. Full text: https://materias.df.uba.ar/f1a2018c2/files/2012/07/falling-cat.pdf (accessed 1 Nov 2026)
NASA Technical Reports Server, record for Kane and Scher (1 July 1969). https://ntrs.nasa.gov/citations/19690058843 (accessed 1 Nov 2026)
É.-J. Marey, "Des mouvements que certains animaux exécutent pour retomber sur leurs pieds lorsqu'ils sont précipités d'un lieu élevé", Comptes rendus hebdomadaires des séances de l'Académie des sciences 119 (1894), pp. 714–717. Catalogue record, Université Paris Cité: https://numerabilis.u-paris.fr/ressources/pdf/medica/bibnum/marey176/marey176.pdf (bibliographic details confirmed; article text not read; accessed 1 Nov 2026)
"Photographs of a tumbling cat", Nature 51 (1308) (1894), pp. 80–81. doi:10.1038/051080a0
The Public Domain Review, "Photographs of a Falling Cat (1894)". https://publicdomainreview.org/collections/photographs-of-a-falling-cat-1894/ (accessed 1 Nov 2026)
Wikipedia, "Falling cat problem" (cross-check only). https://en.wikipedia.org/wiki/Falling_cat_problem (accessed 1 Nov 2026)
